CN112035637A - Medical field intention recognition method, device, equipment and storage medium - Google Patents
Medical field intention recognition method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112035637A CN112035637A CN202010885018.XA CN202010885018A CN112035637A CN 112035637 A CN112035637 A CN 112035637A CN 202010885018 A CN202010885018 A CN 202010885018A CN 112035637 A CN112035637 A CN 112035637A
- Authority
- CN
- China
- Prior art keywords
- target
- preset
- sentence
- medical
- medical field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000000605 extraction Methods 0.000 claims abstract description 46
- 238000013136 deep learning model Methods 0.000 claims abstract description 18
- 238000010606 normalization Methods 0.000 claims description 32
- 238000012545 processing Methods 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 27
- 238000002372 labelling Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract 1
- 208000007882 Gastritis Diseases 0.000 description 18
- 239000013598 vector Substances 0.000 description 12
- 239000003814 drug Substances 0.000 description 9
- 206010046914 Vaginal infection Diseases 0.000 description 7
- 201000008100 Vaginitis Diseases 0.000 description 7
- 238000010276 construction Methods 0.000 description 7
- 229940079593 drug Drugs 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 239000000284 extract Substances 0.000 description 5
- 208000005577 Gastroenteritis Diseases 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 230000004927 fusion Effects 0.000 description 4
- 238000002604 ultrasonography Methods 0.000 description 4
- 208000030507 AIDS Diseases 0.000 description 3
- 239000004083 gastrointestinal agent Substances 0.000 description 3
- 238000009597 pregnancy test Methods 0.000 description 3
- 208000019695 Migraine disease Diseases 0.000 description 2
- 206010037660 Pyrexia Diseases 0.000 description 2
- 208000007107 Stomach Ulcer Diseases 0.000 description 2
- LSQZJLSUYDQPKJ-NJBDSQKTSA-N amoxicillin Chemical compound C1([C@@H](N)C(=O)N[C@H]2[C@H]3SC([C@@H](N3C2=O)C(O)=O)(C)C)=CC=C(O)C=C1 LSQZJLSUYDQPKJ-NJBDSQKTSA-N 0.000 description 2
- 229960003022 amoxicillin Drugs 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 206010012601 diabetes mellitus Diseases 0.000 description 2
- 208000002173 dizziness Diseases 0.000 description 2
- 201000005917 gastric ulcer Diseases 0.000 description 2
- 230000035876 healing Effects 0.000 description 2
- 210000004185 liver Anatomy 0.000 description 2
- 206010027599 migraine Diseases 0.000 description 2
- LSQZJLSUYDQPKJ-UHFFFAOYSA-N p-Hydroxyampicillin Natural products O=C1N2C(C(O)=O)C(C)(C)SC2C1NC(=O)C(N)C1=CC=C(O)C=C1 LSQZJLSUYDQPKJ-UHFFFAOYSA-N 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 210000002784 stomach Anatomy 0.000 description 2
- 230000009747 swallowing Effects 0.000 description 2
- VTYYLEPIZMXCLO-UHFFFAOYSA-L Calcium carbonate Chemical class [Ca+2].[O-]C([O-])=O VTYYLEPIZMXCLO-UHFFFAOYSA-L 0.000 description 1
- 241000238557 Decapoda Species 0.000 description 1
- 206010013700 Drug hypersensitivity Diseases 0.000 description 1
- 208000026072 Motor neurone disease Diseases 0.000 description 1
- 208000025865 Ulcer Diseases 0.000 description 1
- 206010002026 amyotrophic lateral sclerosis Diseases 0.000 description 1
- 229940069428 antacid Drugs 0.000 description 1
- 239000003159 antacid agent Substances 0.000 description 1
- 230000001458 anti-acid effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002496 gastric effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 206010025482 malaise Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 208000005264 motor neuron disease Diseases 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 201000005354 penicillin allergy Diseases 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 231100000397 ulcer Toxicity 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Human Computer Interaction (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses a medical field intention identification method, a device, equipment and a storage medium, which are applied to the field of intelligent medical treatment and used for accurately identifying medical field intention of a user and improving the accuracy of medical field intention identification. The method comprises the following steps: acquiring initial text data, wherein the initial text data comprises question sentences and/or question words input by a target user; calling a preset combined model to rewrite the initial text data to generate a target problem sentence, wherein the target problem sentence is a sentence with clear semantics and no redundancy; performing feature extraction on target question sentences based on a preset medical special feature word stock to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structured information; and calling a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify the plurality of target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
Description
Technical Field
The present invention relates to the field of medical data, and in particular, to a method, an apparatus, a device, and a storage medium for medical field intention recognition.
Background
The intention recognition is an important component of intelligent question answering, namely, the questions of the user are classified into corresponding medical intention categories through a classification method. Simply the intent is the user's will, i.e. what the user wants to do, classified into previously defined intent categories according to the domain and intent that the user expresses. The accuracy of intent recognition is related to the performance of semantic slot filling and facilitates subsequent intelligent question and answer research.
The existing scheme mainly extracts features through a fractal algorithm, key features of a predicted text need to be extracted, and intention features such as characters, words and the like are obtained through extraction.
However, in the medical question and answer field, the expression of the user is not standard, the description is short, and the user has certain ambiguity, so that the user is difficult to acquire rich intention characteristics, and the medical intention identification result of the user is inaccurate.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for identifying medical field intentions, which are used for accurately identifying medical field intentions consulted by a user and improving the accuracy of medical field intention identification results.
A first aspect of an embodiment of the present invention provides a method for identifying medical field intentions, including: acquiring initial text data, wherein the initial text data comprises question sentences and/or question words input by a target user; calling a preset combined model to rewrite the initial text data to generate a target question sentence, wherein the target question sentence is a sentence with clear semantics and no redundancy; extracting the features of the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information; and calling a preset deep learning model, the target feature weights and the target structural information to classify the target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
Optionally, in a first implementation manner of the first aspect of the embodiment of the present invention, the invoking a preset combination model to rewrite the initial text data to generate a target question sentence, where the target question sentence is a sentence with clear semantics and no redundancy, includes: calling a preset combined model to perform entity extraction on the initial text data and perform normalization processing to obtain a plurality of standard entities; searching attribute information corresponding to each standard entity in a preset medical field knowledge graph to obtain a plurality of standard attribute information, wherein each standard attribute information corresponds to a different standard entity; completing the information of the initial text data according to the standard attribute information and the standard entities to generate a plurality of candidate question sentences, and sending the candidate question sentences to a terminal; and determining a target question sentence according to the instruction fed back by the terminal, wherein the target question sentence is a sentence with clear semantics and no redundancy.
Optionally, in a second implementation manner of the first aspect of the embodiment of the present invention, the invoking a preset combination model to perform entity extraction and normalization on the initial text data to obtain a plurality of standard entities includes: calling a preset combined model to perform entity extraction on the initial text data to obtain a plurality of initial entities; and calling a preset normalization knowledge base to carry out normalization processing on the plurality of initial entities to obtain a plurality of standard entities.
Optionally, in a third implementation manner of the first aspect of the embodiment of the present invention, the performing information completion on the initial text data according to the multiple standard attribute information and the multiple standard entities to generate multiple candidate question sentences, and sending the multiple candidate question sentences to the terminal includes: matching the initial text data with a preset sentence pattern template to obtain a target sentence pattern template; adjusting the sentence pattern of the initial text data according to the target sentence pattern template to obtain an initial sentence; replacing a plurality of initial entities in the initial sentence with a plurality of corresponding standard entities to generate an intermediate sentence; completing the intermediate sentences according to the standard attribute information to generate a plurality of candidate question sentences; and sending the candidate question sentences to a terminal.
Optionally, in a fourth implementation manner of the first aspect of the embodiment of the present invention, the performing feature extraction on the target question statement based on a preset medical proprietary feature thesaurus to obtain a plurality of target features, a plurality of target feature weights, and a plurality of target structured information includes: calling the preset combined model to extract entity features of the target question sentences to obtain a plurality of candidate features; calling a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of normalized features; querying contexts corresponding to the plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight; carrying out structuring processing on the corresponding context to obtain a plurality of target structured information; and screening the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
Optionally, in a fifth implementation manner of the first aspect of the embodiment of the present invention, before the obtaining initial text data, where the initial text data includes a question sentence and/or a question word input by a target user, the method for recognizing the medical field intention further includes: and constructing a preset medical knowledge map.
Optionally, in a sixth implementation manner of the first aspect of the embodiment of the present invention, the constructing a preset medical knowledge map includes: acquiring medical training data and integrating the medical training data; performing entity extraction and relation extraction on the medical training data to generate a knowledge triple; carrying out data annotation on the knowledge triple to obtain annotated data; and training and reasoning knowledge on a preset sequence labeling model based on the labeled data to generate a preset medical knowledge map.
A second aspect of an embodiment of the present invention provides a medical field intention identifying apparatus, including: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring initial text data, and the initial text data comprises question sentences and/or question words input by a target user; the data rewriting module is used for calling a preset combined model to rewrite the initial text data and generate a target question sentence, wherein the target question sentence is a sentence with clear semantics and no redundancy; the characteristic extraction module is used for extracting the characteristics of the target question sentences based on a preset medical special characteristic word stock to obtain a plurality of target characteristics, a plurality of target characteristic weights and a plurality of target structural information; and the intention determining module is used for calling a preset deep learning model, the target feature weights and the target structural information to classify the target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
Optionally, in a first implementation manner of the second aspect of the embodiment of the present invention, the data rewriting module includes: the extraction unit is used for calling a preset combined model to perform entity extraction on the initial text data and perform normalization processing to obtain a plurality of standard entities; the searching unit is used for searching the attribute information corresponding to each standard entity in a preset medical field knowledge graph to obtain a plurality of standard attribute information, and each standard attribute information corresponds to a different standard entity; the completion unit is used for performing information completion on the initial text data according to the plurality of standard attribute information and the plurality of standard entities to generate a plurality of candidate question sentences, and sending the candidate question sentences to a terminal; and the determining unit is used for determining a target question sentence according to the instruction fed back by the terminal, wherein the target question sentence is a sentence with clear semantics and no redundancy.
Optionally, in a second implementation manner of the second aspect of the embodiment of the present invention, the extracting unit is specifically configured to: calling a preset combined model to perform entity extraction on the initial text data to obtain a plurality of initial entities; and calling a preset normalization knowledge base to carry out normalization processing on the plurality of initial entities to obtain a plurality of standard entities.
Optionally, in a third implementation manner of the second aspect of the embodiment of the present invention, the completion unit is specifically configured to: matching the initial text data with a preset sentence pattern template to obtain a target sentence pattern template; adjusting the sentence pattern of the initial text data according to the target sentence pattern template to obtain an initial sentence; replacing a plurality of initial entities in the initial sentence with a plurality of corresponding standard entities to generate an intermediate sentence; completing the intermediate sentences according to the standard attribute information to generate a plurality of candidate question sentences; and sending the candidate question sentences to a terminal.
Optionally, in a fourth implementation manner of the second aspect of the embodiment of the present invention, the feature extraction module is specifically configured to: calling the preset combined model to extract entity features of the target question sentences to obtain a plurality of candidate features; calling a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of normalized features; querying contexts corresponding to the plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight; carrying out structuring processing on the corresponding context to obtain a plurality of target structured information; and screening the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
Optionally, in a fifth implementation manner of the second aspect of the embodiment of the present invention, the medical field intention identifying apparatus further includes: and the map construction module is used for constructing a preset medical knowledge map.
Optionally, in a sixth implementation manner of the second aspect of the embodiment of the present invention, the map building module is specifically configured to: acquiring medical training data and integrating the medical training data; performing entity extraction and relation extraction on the medical training data to generate a knowledge triple; carrying out data annotation on the knowledge triple to obtain annotated data; and training and reasoning knowledge on a preset sequence labeling model based on the labeled data to generate a preset medical knowledge map.
A third aspect of embodiments of the present invention provides a medical field intention identifying apparatus, a memory having instructions stored therein and at least one processor, the memory and the at least one processor being interconnected by a line; the at least one processor invokes the instructions in the memory to cause the medical field intent recognition device to perform the medical field intent recognition method described above.
A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the medical field intention identifying method according to any one of the above embodiments.
In the technical scheme provided by the embodiment of the invention, initial text data is obtained, and the initial text data comprises problem sentences and/or problem words input by a target user; calling a preset combined model to rewrite the initial text data to generate a target problem sentence, wherein the target problem sentence is a sentence with clear semantics and no redundancy; performing feature extraction on target question sentences based on a preset medical special feature word stock to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structured information; and calling a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify the plurality of target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result. According to the embodiment of the invention, the medical field intention of the user can be accurately identified, and the accuracy of the medical field intention identification is improved.
Drawings
FIG. 1 is a diagram of an embodiment of a method for identifying medical field intentions in an embodiment of the present invention;
FIG. 2 is a schematic diagram of another embodiment of a method for identifying medical field intentions in an embodiment of the present invention;
FIG. 3 is a schematic diagram of an embodiment of a medical field intention recognition apparatus in an embodiment of the invention;
FIG. 4 is a schematic diagram of another embodiment of a medical field intention identifying apparatus according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an embodiment of a medical field intention recognition device in an embodiment of the invention.
Detailed Description
The invention provides a medical field intention recognition method, a medical field intention recognition device and a storage medium, which are used for accurately recognizing medical field intentions of users and improving the accuracy of medical field intention recognition.
In order to make the technical field of the invention better understand the scheme of the invention, the embodiment of the invention will be described in conjunction with the attached drawings in the embodiment of the invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Referring to fig. 1, a flowchart of a method for identifying an intention in a medical field according to an embodiment of the present invention specifically includes:
101. initial text data is obtained, the initial text data including question sentences and/or question words input by a target user.
The server obtains initial text data including question sentences and/or question words input by the target user. The question sentence may include multiple forms, and the question sentence may be a sentence with clear semantics and simple expression, for example: "how to treat gastroenteritis? "," what medicine is needed for gastroenteritis? "," what do children need to get a fever? "and the like; the question sentence may also be a semantically ambiguous sentence, such as: "you help me see", "what do i, feeling uncomfortable", "feeling dizziness in general", etc.; the question sentence may also be a sentence expressing redundancy, for example, "my stomach is somewhat uncomfortable, body is uncomfortable", "i throat is somewhat uncomfortable, swallowing difficulty, may be getting inflamed", "yesterday goes to do liver color ultrasound, B ultrasound results are echo-dense, enhanced, thickened", the question word may include only one keyword, for example, the question word may be "diabetes", "migraine", "gastritis", "vaginitis", etc.
It is to be understood that the executing subject of the present invention may be the medical field intention identifying apparatus, and may also be a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
102. And calling a preset combined model to rewrite the initial text data to generate a target question sentence, wherein the target question sentence is a sentence with clear semantics and no redundancy.
And the server calls a preset combined model to rewrite the initial text data to generate a target problem sentence, wherein the target problem sentence is a sentence with clear semantics and no redundancy. Specifically, the server calls a preset combined model to perform entity extraction and normalization processing on initial text data to obtain a plurality of standard entities; the server searches attribute information corresponding to each standard entity in a preset medical field knowledge graph to obtain a plurality of standard attribute information, wherein each standard attribute information corresponds to a different standard entity; the server completes the information of the initial text data according to the plurality of standard attribute information and the plurality of standard entities to generate a plurality of candidate question sentences and sends the plurality of candidate question sentences to the terminal; and the server determines a target problem statement according to the instruction fed back by the terminal, wherein the target problem statement is a sentence with clear semantics and no redundancy.
For example, the user inputs "gastritis", and cannot understand that the user wants to consult a specific problem about "gastritis", and searches for the attribute of "gastritis" through the knowledge graph in the medical field to generate two candidate problem sentences "how to treat gastritis? "and" what need to be noticed with gastritis? Sending the two candidate question sentences to a terminal used by a target user so that the target user can select on the terminal, and determining the target question sentences when acquiring instructions fed back by the terminal. If, the target user selected "what treatment did gastritis? ", then the question statement is the target question statement. It can be understood that the selection of the user is more beneficial to understanding the requirements of the user.
103. And performing feature extraction on the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information.
The server extracts features of the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information. Specifically, the server calls a preset combination model to extract entity features of the target question sentences to obtain a plurality of candidate features; the server calls a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of standardized features; the server inquires context corresponding to a plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight; the server carries out structuralization processing on the corresponding context to obtain a plurality of target structuralization information; the server screens the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
The target structured information may be a sentence word vector feature, a position vector, a manually extracted feature or an entity weighted feature, and may further include other structured information, which is not limited herein.
It should be noted that, based on the normative and rigor of the professional medical text description, a medical special feature word library, such as a drug library, a disease library, a treatment library, etc., is constructed in advance. The medical-specific feature thesaurus can be used for improving the accuracy of the score segmentation and retrieving related information of the medical entity.
104. And calling a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify the plurality of target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
The server calls a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify a plurality of target features to obtain a classification result, and the medical field intention of the target user is determined according to the classification result.
The deep learning model uses an improved textcnn model, an input layer comprises sentence word vector characteristics, position vectors, manually extracted characteristics and entity weighted characteristics, characteristic fusion is carried out to obtain input layer characteristics (fusion characteristics), the input layer characteristics are transmitted into a convolutional layer, the sentence characteristic vectors are learned through the convolutional layer, intention classification is completed through a full connection layer by using the sentence characteristic vectors, and the size of the sliding step length of a characteristic graph of the convolutional layer is set according to the average length of a medical information base constructed by training data entities and the characteristics of medical texts.
It should be noted that the preset deep learning model is a model trained in advance, in order to improve the stability of the model, data enhancement is performed on limited labeled data, the existing learning model is used to perform pre-marking on unsupervised data, data with high confidence coefficient is selected and added into the labeled data, and the enhanced labeled data, namely training data, is obtained, so that the enhancement of the training data is realized.
According to the embodiment of the invention, the deep learning is combined with the special characteristics of the medical text, the text input by the user is rewritten, the medical intention recognition is carried out on the rewritten text by combining the knowledge map of the medical field, the medical field intention consulted by the user can be accurately recognized, and the accuracy of the intention recognition result of the medical field is improved. And this scheme can be applied to in the wisdom medical treatment field to promote the construction in wisdom city.
Referring to fig. 2, another flowchart of the method for identifying medical field intention according to the embodiment of the present invention specifically includes:
201. initial text data is obtained, the initial text data including question sentences and/or question words input by a target user.
The server obtains initial text data including question sentences and/or question words input by the target user. The question sentence may include multiple forms, and the question sentence may be a sentence with clear semantics and simple expression, for example: "how to treat gastroenteritis? "," what medicine is needed for gastroenteritis? "," what do children need to get a fever? "and the like; the question sentence may also be a semantically ambiguous sentence, such as: "you help me see", "what do i, feeling uncomfortable", "feeling dizziness in general", etc.; the question sentence may also be a sentence expressing redundancy, for example, "my stomach is somewhat uncomfortable, body is uncomfortable", "i throat is somewhat uncomfortable, swallowing difficulty, may be getting inflamed", "yesterday goes to do liver color ultrasound, B ultrasound results are echo-dense, enhanced, thickened", the question word may include only one keyword, for example, the question word may be "diabetes", "migraine", "gastritis", "vaginitis", etc.
It is to be understood that the executing subject of the present invention may be the medical field intention identifying apparatus, and may also be a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
Optionally, before step 201, the method may further include:
the server constructs a preset medical knowledge map.
Optionally, the server specifically constructs the preset medical knowledge graph, including:
the server acquires medical training data and integrates the medical training data; the server performs entity extraction and relation extraction on the medical training data to generate a knowledge triple; the server carries out data annotation on the knowledge triple to obtain annotated data; and the server trains and reasoning knowledge on the preset sequence labeling model based on the labeled data to generate a preset medical knowledge map.
Optionally, the server performs entity extraction and relationship extraction on the medical training data to generate a knowledge triple, including: the server performs entity extraction on the medical training data to obtain a plurality of entities; the server extracts the relationships among the entities to obtain a plurality of entity relationships; the server generates a knowledge triple according to the plurality of entity relationships and the plurality of entities.
The entity identification refers to identifying entities with specific meanings in medical texts, and mainly comprises entities such as subjects, places, inspection items, diseases and the like, and entity relation extraction needs to infer the relation among the entities according to text semantic information; such as given the sentence: the entity is the pregnant woman and the crab, the preset combined extraction model can obtain the unsuitable relation through semantics, and finally extract the knowledge triple of the pregnant woman, unsuitable crabs. Knowledge triples are related entity triples.
The preset combined model is used for extracting the relation between the entity and the entity, and the labeled data of a BIOES mode is adopted, wherein B represents the beginning, I represents the inside, O represents a non-entity, E represents the tail of the entity, and S represents that the word change is an entity. The serialization labeling model is a BILSTM + CRF deep learning model, and the BILSTM + CRF deep learning model is trained and optimized by using labeled data (labeled triples) to obtain the medical knowledge map.
202. Calling a preset combined model to perform entity extraction and normalization processing on the initial text data to obtain a plurality of standard entities;
specifically, the server calls a preset combined model to perform entity extraction on initial text data to obtain a plurality of initial entities; and the server calls a preset normalization knowledge base to carry out normalization processing on the plurality of initial entities to obtain a plurality of standard entities.
For example, "pregnancy test paper" and "pregnancy test paper" are both standardized as "pregnancy test paper". As another example, "motor neuron disease" and "summer sickness" are standardized as "amyotrophic lateral sclerosis", "AIDS" and "AIDS" are standardized as "acquired immunodeficiency syndrome".
It should be noted that entity extraction and normalization are performed on the initial text data of the user, and the normalization of the user text is helpful for subsequent intention recognition, thereby avoiding semantic inaccuracy and redundancy.
203. Searching attribute information corresponding to each standard entity in a preset medical field knowledge graph to obtain a plurality of standard attribute information, wherein each standard attribute information corresponds to a different standard entity;
different standard entities correspond to different standard attribute information, for example, attribute information corresponding to "gastritis" is "treatment method" and "attention" and attribute information corresponding to "amoxicillin" is "indication" and "contraindication".
204. Performing information completion on the initial text data according to the plurality of standard attribute information and the plurality of standard entities to generate a plurality of candidate question sentences, and sending the plurality of candidate question sentences to the terminal;
specifically, the server matches the initial text data with a preset sentence pattern template to obtain a target sentence pattern template; the server adjusts the sentence pattern of the initial text data according to the target sentence pattern template to obtain an initial sentence; the server replaces a plurality of initial entities in the initial sentences with a plurality of corresponding standard entities to generate intermediate sentences; the server completes the intermediate sentences according to the standard attribute information to generate a plurality of candidate question sentences; the server sends the plurality of candidate question sentences to the terminal.
For example, the initial text data is "what different colors are for the B-mode and color-mode," what difference is between a and B "sentence pattern is matched by a preset sentence pattern template, the template corresponding to the sentence pattern is" (. about.) ([ and with ]) (. about.)) - {0} { and } {2} { has difference } ", and" what difference is between the B-mode and the color-mode "is finally modified to" what difference is between the B-mode and the color-mode "according to the preset sentence pattern template.
For example, the user inputs "vaginitis", the corresponding standard entity is also "vaginitis", the attribute information of "vaginitis" is found to be "treatment method" and "notice" through the knowledge map, and then the user is asked the question "how to treat vaginitis? What is the vaginitis required to be noticed? "
The attribute information is an attribute list constructed for the entity, the attribute of the drug includes indication and contraindication, and the attribute of the disease may include treatment medication, caution, and the like, for example, amoxicillin is forbidden by people with penicillin allergy.
For example, the user inputs "gastritis", and cannot understand that the user wants to consult a specific problem about "gastritis", and searches for the attribute of "gastritis" through the knowledge graph in the medical field to generate two candidate problem sentences "how to treat gastritis? "and" what need to be noticed with gastritis? Sending the two candidate question sentences to a terminal used by a target user so that the target user can select on the terminal, and determining the target question sentences when acquiring instructions fed back by the terminal.
205. Determining a target problem statement according to an instruction fed back by a terminal, wherein the target problem statement is a sentence with clear semantics and no redundancy;
for example, for the question "how to treat gastritis? "and" what need to be noticed with gastritis? ", the target user has selected" how to treat gastritis? "then the question sentence" how to treat gastritis? "determined as a target question statement. It can be understood that the selection of the target user is more beneficial to understanding the requirements of the user.
206. And performing feature extraction on the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information.
The server extracts features of the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information. Specifically, the server calls a preset combination model to extract entity features of the target question sentences to obtain a plurality of candidate features; the server calls a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of standardized features; the server inquires context corresponding to a plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight; the server carries out structuralization processing on the corresponding context to obtain a plurality of target structuralization information; the server screens the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
The target structured information may be a sentence word vector feature, a position vector, a manually extracted feature or an entity weighted feature, and may further include other structured information, which is not limited herein.
It should be noted that, based on the normative and rigor of the professional medical text description, a medical special feature word library, such as a drug library, a disease library, a treatment library, etc., is constructed in advance. The medical-specific feature thesaurus can be used for improving the accuracy of the score segmentation and retrieving related information of the medical entity.
207. And calling a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify the plurality of target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
The server calls a preset deep learning model, a plurality of target feature weights and a plurality of target structural information to classify a plurality of target features to obtain a classification result, and the medical field intention of the target user is determined according to the classification result.
The deep learning model uses an improved textcnn model, an input layer comprises sentence word vector characteristics, position vectors, manually extracted characteristics and entity weighted characteristics, characteristic fusion is carried out to obtain input layer characteristics (fusion characteristics), the input layer characteristics are transmitted into a convolutional layer, the sentence characteristic vectors are learned through the convolutional layer, intention classification is completed through a full connection layer by using the sentence characteristic vectors, and the size of the sliding step length of a characteristic graph of the convolutional layer is set according to the average length of a medical information base constructed by training data entities and the characteristics of medical texts.
It should be noted that the preset deep learning model is a model trained in advance, in order to improve the stability of the model, data enhancement is performed on limited labeled data, the existing learning model is used to perform pre-marking on unsupervised data, data with high confidence coefficient is selected and added into the labeled data, and the enhanced labeled data, namely training data, is obtained, so that the enhancement of the training data is realized.
208. And searching answers to the questions according to the medical field intention of the target user.
The server searches answers to the questions according to the medical field intention of the target user.
Optionally, when the medical field of the target user is intended to query a drug for treating gastric ulcer, the server determines the relevant information of the gastric drug in a preset medical knowledge map; the server classifies the related information of the gastric drugs to obtain the information of the antacid gastric drugs and the information of the ulcer healing gastric drugs; and the server generates a question answer according to the information of the gastric ulcer healing medicines in a preset format and feeds the question answer back to a target terminal of a target user.
Because the medical field intention in the user input question can be accurately identified, the answer required by the user input question can be more accurately retrieved, the retrieval time of the answer is shortened, and the answer retrieval efficiency is improved.
According to the embodiment of the invention, the deep learning is combined with the special characteristics of the medical text, the text input by the user is rewritten, the medical intention recognition is carried out on the rewritten text by combining the knowledge map of the medical field, the medical field intention consulted by the user can be accurately recognized, and the accuracy of the intention recognition result of the medical field is improved. And this scheme can be applied to in the wisdom medical treatment field to promote the construction in wisdom city.
In the above description of the method for recognizing a medical field intention according to the embodiment of the present invention, referring to fig. 3, a medical field intention recognition apparatus according to the embodiment of the present invention is described below, and an embodiment of the medical field intention recognition apparatus according to the embodiment of the present invention includes:
a data obtaining module 301, configured to obtain initial text data, where the initial text data includes question sentences and/or question words input by a target user;
a data rewriting module 302, configured to call a preset combination model to rewrite the initial text data, and generate a target question sentence, where the target question sentence is a sentence with clear semantics and no redundancy;
a feature extraction module 303, configured to perform feature extraction on the target question statement based on a preset medical special feature lexicon to obtain multiple target features, multiple target feature weights, and multiple target structured information;
an intention determining module 304, configured to invoke a preset deep learning model, the multiple target feature weights, and the multiple target structural information to classify the multiple target features, obtain a classification result, and determine a medical field intention of the target user according to the classification result.
According to the embodiment of the invention, the deep learning is combined with the special characteristics of the medical text, the text input by the user is rewritten, the medical intention recognition is carried out on the rewritten text by combining the knowledge map of the medical field, the medical field intention consulted by the user can be accurately recognized, and the accuracy of the intention recognition result of the medical field is improved. And this scheme can be applied to in the wisdom medical treatment field to promote the construction in wisdom city.
Referring to fig. 4, another embodiment of the device for recognizing medical field intention according to the embodiment of the present invention includes:
a data obtaining module 301, configured to obtain initial text data, where the initial text data includes question sentences and/or question words input by a target user;
a data rewriting module 302, configured to call a preset combination model to rewrite the initial text data, and generate a target question sentence, where the target question sentence is a sentence with clear semantics and no redundancy;
a feature extraction module 303, configured to perform feature extraction on the target question statement based on a preset medical special feature lexicon to obtain multiple target features, multiple target feature weights, and multiple target structured information;
an intention determining module 304, configured to invoke a preset deep learning model, the multiple target feature weights, and the multiple target structural information to classify the multiple target features, obtain a classification result, and determine a medical field intention of the target user according to the classification result.
Optionally, the data rewriting module 302 includes:
an extraction unit 3021, configured to invoke a preset combination model to perform entity extraction on the initial text data and perform normalization processing to obtain multiple standard entities;
the searching unit 3022 is configured to search the attribute information corresponding to each standard entity in a preset medical domain knowledge graph to obtain a plurality of standard attribute information, where each standard attribute information corresponds to a different standard entity;
a completion unit 3023, configured to perform information completion on the initial text data according to the plurality of standard attribute information and the plurality of standard entities, generate a plurality of candidate question sentences, and send the plurality of candidate question sentences to a terminal;
a determining unit 3024, configured to determine a target question statement according to the instruction fed back by the terminal, where the target question statement is a sentence with clear semantics and no redundancy.
Optionally, the extracting unit 3021 is specifically configured to:
calling a preset combined model to perform entity extraction on the initial text data to obtain a plurality of initial entities; and calling a preset normalization knowledge base to carry out normalization processing on the plurality of initial entities to obtain a plurality of standard entities.
Optionally, the completion unit 3023 is specifically configured to:
matching the initial text data with a preset sentence pattern template to obtain a target sentence pattern template; adjusting the sentence pattern of the initial text data according to the target sentence pattern template to obtain an initial sentence; replacing a plurality of initial entities in the initial sentence with a plurality of corresponding standard entities to generate an intermediate sentence; completing the intermediate sentences according to the standard attribute information to generate a plurality of candidate question sentences; and sending the candidate question sentences to a terminal.
Optionally, the feature extraction module 303 is specifically configured to:
calling the preset combined model to extract entity features of the target question sentences to obtain a plurality of candidate features; calling a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of normalized features; querying contexts corresponding to the plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight; carrying out structuring processing on the corresponding context to obtain a plurality of target structured information; and screening the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
Optionally, the medical field intention identifying apparatus further comprises:
and the map construction module 305 is used for constructing a preset medical knowledge map.
Optionally, the map building module 305 is specifically configured to:
acquiring medical training data and integrating the medical training data; performing entity extraction and relation extraction on the medical training data to generate a knowledge triple; carrying out data annotation on the knowledge triple to obtain annotated data; and training and reasoning knowledge on a preset sequence labeling model based on the labeled data to generate a preset medical knowledge map.
According to the embodiment of the invention, the deep learning is combined with the special characteristics of the medical text, the text input by the user is rewritten, the medical intention recognition is carried out on the rewritten text by combining the knowledge map of the medical field, the medical field intention consulted by the user can be accurately recognized, and the accuracy of the intention recognition result of the medical field is improved. And this scheme can be applied to in the wisdom medical treatment field to promote the construction in wisdom city.
Fig. 3 to 4 describe the medical field intention recognition apparatus in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the medical field intention recognition apparatus in the embodiment of the present invention is described in detail from the perspective of the hardware processing.
Fig. 5 is a schematic structural diagram of a medical field intention recognition apparatus 500 according to an embodiment of the present invention, which may have relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing applications 533 or data 532. Memory 520 and storage media 530 may be, among other things, transient or persistent storage. The program stored on the storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations for the medical field intention recognition apparatus 500. Still further, the processor 510 may be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the medical field intention recognition device 500.
The medical field intent recognition device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input-output interfaces 560, and/or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that the medical field intention recognition apparatus configuration shown in fig. 5 does not constitute a limitation of the medical field intention recognition apparatus, and may include more or less components than those shown, or combine some components, or a different arrangement of components. The processor 510 may perform the functions of the data acquisition module 301, the data rewriting module 302, the feature extraction module 303, the intention determination module 304, and the atlas construction module 305 in the embodiments described above.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, which may also be a volatile computer readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the medical field intention identifying method.
The present invention also provides a medical field intention recognition apparatus, which includes a memory and a processor, wherein the memory stores instructions, and the instructions, when executed by the processor, cause the processor to execute the steps of the medical field intention recognition method in the above embodiments.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A medical field intention recognition method, comprising:
acquiring initial text data, wherein the initial text data comprises question sentences and/or question words input by a target user;
calling a preset combined model to rewrite the initial text data to generate a target question sentence, wherein the target question sentence is a sentence with clear semantics and no redundancy;
extracting the features of the target question sentences based on a preset medical special feature word bank to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structural information;
and calling a preset deep learning model, the target feature weights and the target structural information to classify the target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
2. The method for recognizing the medical field intention according to claim 1, wherein the step of calling a preset joint model to rewrite the initial text data to generate a target question sentence, wherein the target question sentence is a semantically clear and non-redundant sentence, and comprises the steps of:
calling a preset combined model to perform entity extraction on the initial text data and perform normalization processing to obtain a plurality of standard entities;
searching attribute information corresponding to each standard entity in a preset medical field knowledge graph to obtain a plurality of standard attribute information, wherein each standard attribute information corresponds to a different standard entity;
completing the information of the initial text data according to the standard attribute information and the standard entities to generate a plurality of candidate question sentences, and sending the candidate question sentences to a terminal;
and determining a target question sentence according to the instruction fed back by the terminal, wherein the target question sentence is a sentence with clear semantics and no redundancy.
3. The method for recognizing the medical field intention according to claim 2, wherein the step of calling a preset combination model to perform entity extraction and normalization processing on the initial text data to obtain a plurality of standard entities comprises:
calling a preset combined model to perform entity extraction on the initial text data to obtain a plurality of initial entities;
and calling a preset normalization knowledge base to carry out normalization processing on the plurality of initial entities to obtain a plurality of standard entities.
4. The method for recognizing medical field intention according to claim 3, wherein the performing information completion on the initial text data according to the plurality of standard attribute information and the plurality of standard entities to generate a plurality of candidate question sentences, and sending the plurality of candidate question sentences to a terminal includes:
matching the initial text data with a preset sentence pattern template to obtain a target sentence pattern template;
adjusting the sentence pattern of the initial text data according to the target sentence pattern template to obtain an initial sentence;
replacing a plurality of initial entities in the initial sentence with a plurality of corresponding standard entities to generate an intermediate sentence;
completing the intermediate sentences according to the standard attribute information to generate a plurality of candidate question sentences;
and sending the candidate question sentences to a terminal.
5. The method for recognizing the medical field intention according to claim 1, wherein the extracting the target question sentence based on the preset medical specific feature lexicon to obtain a plurality of target features, a plurality of target feature weights and a plurality of target structured information comprises:
calling the preset combined model to extract entity features of the target question sentences to obtain a plurality of candidate features;
calling a preset normalization knowledge base to carry out normalization processing on the candidate features to obtain a plurality of normalized features;
querying contexts corresponding to the plurality of standardized features and a plurality of target feature weights from a preset medical field knowledge graph, wherein each standardized feature corresponds to one target feature weight;
carrying out structuring processing on the corresponding context to obtain a plurality of target structured information;
and screening the plurality of standardized features based on a preset medical special feature word bank to obtain a plurality of target features.
6. The medical field intention recognition method of any one of claims 1-5, wherein prior to the obtaining initial text data comprising question sentences and/or question words input by a target user, the medical field intention recognition method further comprises:
and constructing a preset medical knowledge map.
7. The medical field intention recognition method of claim 6, wherein said constructing a preset medical knowledge-graph comprises:
acquiring medical training data and integrating the medical training data;
performing entity extraction and relation extraction on the medical training data to generate a knowledge triple;
carrying out data annotation on the knowledge triple to obtain annotated data;
and training and reasoning knowledge on a preset sequence labeling model based on the labeled data to generate a preset medical knowledge map.
8. A medical field intention recognition apparatus, comprising:
the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring initial text data, and the initial text data comprises question sentences and/or question words input by a target user;
the data rewriting module is used for calling a preset combined model to rewrite the initial text data and generate a target question sentence, wherein the target question sentence is a sentence with clear semantics and no redundancy;
the characteristic extraction module is used for extracting the characteristics of the target question sentences based on a preset medical special characteristic word stock to obtain a plurality of target characteristics, a plurality of target characteristic weights and a plurality of target structural information;
and the intention determining module is used for calling a preset deep learning model, the target feature weights and the target structural information to classify the target features to obtain a classification result, and determining the medical field intention of the target user according to the classification result.
9. A medical field intention recognition apparatus characterized by comprising: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line;
the at least one processor invokes the instructions in the memory to cause the medical field intent recognition device to perform the medical field intent recognition method of any of claims 1-7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores instructions that, when executed by a processor, implement the medical field intent recognition method of any one of claims 1-7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010885018.XA CN112035637A (en) | 2020-08-28 | 2020-08-28 | Medical field intention recognition method, device, equipment and storage medium |
| PCT/CN2021/084653 WO2022041728A1 (en) | 2020-08-28 | 2021-03-31 | Medical field intention recognition method, apparatus, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010885018.XA CN112035637A (en) | 2020-08-28 | 2020-08-28 | Medical field intention recognition method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112035637A true CN112035637A (en) | 2020-12-04 |
Family
ID=73586083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010885018.XA Pending CN112035637A (en) | 2020-08-28 | 2020-08-28 | Medical field intention recognition method, device, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112035637A (en) |
| WO (1) | WO2022041728A1 (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112735543A (en) * | 2020-12-30 | 2021-04-30 | 杭州依图医疗技术有限公司 | Medical data processing method and device and storage medium |
| CN113256459A (en) * | 2021-04-30 | 2021-08-13 | 深圳市鹰硕教育服务有限公司 | Micro-course video management method, device, system and storage medium |
| CN113255354A (en) * | 2021-06-03 | 2021-08-13 | 北京达佳互联信息技术有限公司 | Search intention recognition method, device, server and storage medium |
| CN113642312A (en) * | 2021-08-19 | 2021-11-12 | 平安医疗健康管理股份有限公司 | Physical examination data processing method, physical examination data processing device, physical examination equipment and storage medium |
| CN113657102A (en) * | 2021-08-17 | 2021-11-16 | 北京百度网讯科技有限公司 | Information extraction method, information extraction device, information extraction apparatus, storage medium, and program |
| CN113657114A (en) * | 2021-08-31 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Method, device, equipment and storage medium for generating disease name code matching list |
| CN113761899A (en) * | 2021-09-07 | 2021-12-07 | 卫宁健康科技集团股份有限公司 | A method, device, device and storage medium for generating medical text |
| CN113886400A (en) * | 2021-10-26 | 2022-01-04 | 卫宁健康科技集团股份有限公司 | Medical data management method and device, computer equipment and storage medium |
| WO2022041728A1 (en) * | 2020-08-28 | 2022-03-03 | 康键信息技术（深圳）有限公司 | Medical field intention recognition method, apparatus, device and storage medium |
| CN114724078A (en) * | 2022-03-28 | 2022-07-08 | 西南交通大学 | Human behavior intention recognition method based on target detection network and knowledge reasoning |
| CN115146037A (en) * | 2021-08-09 | 2022-10-04 | 上海蓬海涞讯数据技术有限公司 | Knowledge graph-based question and answer method and system, electronic equipment and storage medium |
| CN115862896A (en) * | 2023-02-13 | 2023-03-28 | 深圳市汇健智慧医疗有限公司 | Perioperative period-based doctor-patient cooperative management method and system |
| CN116468396A (en) * | 2023-04-14 | 2023-07-21 | 浪潮智慧科技有限公司 | Water conservancy service processing method, device and medium based on intent analysis model |
| CN116842145A (en) * | 2023-04-20 | 2023-10-03 | 海信集团控股股份有限公司 | Domain identification method and device based on city question-answering system |
| CN117034957A (en) * | 2023-06-30 | 2023-11-10 | 海信集团控股股份有限公司 | Semantic understanding method and device |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114817497B (en) * | 2022-04-13 | 2024-12-20 | 西安电子科技大学 | A hybrid question answering method based on intent recognition and template matching |
| CN114780713A (en) * | 2022-04-26 | 2022-07-22 | 慧择保险经纪有限公司 | Information extraction method and device and electronic equipment |
| CN114925181B (en) * | 2022-04-28 | 2024-10-18 | 支付宝(杭州)信息技术有限公司 | Data processing method and device, computer storage medium and terminal |
| CN114842982B (en) * | 2022-07-06 | 2022-09-23 | 广东省科技基础条件平台中心 | Knowledge expression method, device and system for medical information system |
| CN115270802B (en) * | 2022-09-29 | 2023-01-03 | 中科雨辰科技有限公司 | Question sentence processing method, electronic equipment and storage medium |
| CN116364296B (en) * | 2023-02-17 | 2023-12-26 | 中国人民解放军总医院 | Standard inspection project name confirmation method, device, equipment, medium and product |
| CN116483982B (en) * | 2023-06-25 | 2023-10-13 | 北京中关村科金技术有限公司 | Knowledge question-answering method, knowledge question-answering device, electronic equipment and readable storage medium |
| CN116933800B (en) * | 2023-09-12 | 2024-01-05 | 深圳须弥云图空间科技有限公司 | Template-based generation type intention recognition method and device |
| CN117370596A (en) * | 2023-10-13 | 2024-01-09 | 北京百度网讯科技有限公司 | Medicine knowledge retrieval method and device |
| CN117252211A (en) * | 2023-10-18 | 2023-12-19 | 中建八局第一数字科技有限公司 | Construction of questioning intention recognition model and questioning intention recognition method and device |
| CN117171332A (en) * | 2023-11-02 | 2023-12-05 | 江西拓世智能科技股份有限公司 | Intelligent question-answering method and system based on AI |
| CN117725961B (en) * | 2024-02-18 | 2024-07-30 | 智慧眼科技股份有限公司 | Medical intention recognition model training method, medical intention recognition method and equipment |
| CN117874240B (en) * | 2024-03-12 | 2024-06-14 | 国网天津市电力公司城西供电分公司 | Audit text classification method, system and equipment based on knowledge graph |
| CN118070909B (en) * | 2024-04-25 | 2024-10-01 | 支付宝(杭州)信息技术有限公司 | Answer generation method and device of medical question-answering system |
| CN118132729B (en) * | 2024-04-28 | 2024-08-16 | 支付宝(杭州)信息技术有限公司 | Answer generation method and device based on medical knowledge graph |
| CN118469005B (en) * | 2024-07-10 | 2024-09-27 | 北方健康医疗大数据科技有限公司 | Medical knowledge graph construction method, system, terminal and storage medium based on large language model |
| CN118966237A (en) * | 2024-10-14 | 2024-11-15 | 智慧眼科技股份有限公司 | Entity attribute information determination method, device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522544A (en) * | 2018-09-27 | 2019-03-26 | 厦门快商通信息技术有限公司 | Sentence vector calculation, file classification method and system based on Chi-square Test |
| CN109657238A (en) * | 2018-12-10 | 2019-04-19 | 宁波深擎信息科技有限公司 | Context identification complementing method, system, terminal and the medium of knowledge based map |
| CN109918489A (en) * | 2019-02-28 | 2019-06-21 | 上海乐言信息科技有限公司 | A kind of knowledge question answering method and system of more strategy fusions |
| CN110675944A (en) * | 2019-09-20 | 2020-01-10 | 京东方科技集团股份有限公司 | Triage method and device, computer equipment and medium |
| CN111339777A (en) * | 2020-02-24 | 2020-06-26 | 中国科学院自动化研究所 | Method and system for medical-related intent recognition based on neural network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729493A (en) * | 2017-09-29 | 2018-02-23 | 北京创鑫旅程网络技术有限公司 | Travel the construction method of knowledge mapping, device and travelling answering method, device |
| CN111177328B (en) * | 2018-11-12 | 2023-04-28 | 阿里巴巴集团控股有限公司 | Question-answer matching system and method, question-answer processing device and medium |
| CN112035637A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Medical field intention recognition method, device, equipment and storage medium |
-
2020
- 2020-08-28 CN CN202010885018.XA patent/CN112035637A/en active Pending
-
2021
- 2021-03-31 WO PCT/CN2021/084653 patent/WO2022041728A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522544A (en) * | 2018-09-27 | 2019-03-26 | 厦门快商通信息技术有限公司 | Sentence vector calculation, file classification method and system based on Chi-square Test |
| CN109657238A (en) * | 2018-12-10 | 2019-04-19 | 宁波深擎信息科技有限公司 | Context identification complementing method, system, terminal and the medium of knowledge based map |
| CN109918489A (en) * | 2019-02-28 | 2019-06-21 | 上海乐言信息科技有限公司 | A kind of knowledge question answering method and system of more strategy fusions |
| CN110675944A (en) * | 2019-09-20 | 2020-01-10 | 京东方科技集团股份有限公司 | Triage method and device, computer equipment and medium |
| CN111339777A (en) * | 2020-02-24 | 2020-06-26 | 中国科学院自动化研究所 | Method and system for medical-related intent recognition based on neural network |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022041728A1 (en) * | 2020-08-28 | 2022-03-03 | 康键信息技术（深圳）有限公司 | Medical field intention recognition method, apparatus, device and storage medium |
| CN112735543A (en) * | 2020-12-30 | 2021-04-30 | 杭州依图医疗技术有限公司 | Medical data processing method and device and storage medium |
| CN113256459A (en) * | 2021-04-30 | 2021-08-13 | 深圳市鹰硕教育服务有限公司 | Micro-course video management method, device, system and storage medium |
| CN113255354A (en) * | 2021-06-03 | 2021-08-13 | 北京达佳互联信息技术有限公司 | Search intention recognition method, device, server and storage medium |
| CN115146037A (en) * | 2021-08-09 | 2022-10-04 | 上海蓬海涞讯数据技术有限公司 | Knowledge graph-based question and answer method and system, electronic equipment and storage medium |
| CN113657102A (en) * | 2021-08-17 | 2021-11-16 | 北京百度网讯科技有限公司 | Information extraction method, information extraction device, information extraction apparatus, storage medium, and program |
| CN113657102B (en) * | 2021-08-17 | 2023-05-30 | 北京百度网讯科技有限公司 | Information extraction method, device, equipment and storage medium |
| CN113642312A (en) * | 2021-08-19 | 2021-11-12 | 平安医疗健康管理股份有限公司 | Physical examination data processing method, physical examination data processing device, physical examination equipment and storage medium |
| CN113657114A (en) * | 2021-08-31 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Method, device, equipment and storage medium for generating disease name code matching list |
| CN113761899A (en) * | 2021-09-07 | 2021-12-07 | 卫宁健康科技集团股份有限公司 | A method, device, device and storage medium for generating medical text |
| CN113761899B (en) * | 2021-09-07 | 2024-11-22 | 卫宁健康科技集团股份有限公司 | A medical text generation method, device, equipment and storage medium |
| CN113886400A (en) * | 2021-10-26 | 2022-01-04 | 卫宁健康科技集团股份有限公司 | Medical data management method and device, computer equipment and storage medium |
| CN114724078A (en) * | 2022-03-28 | 2022-07-08 | 西南交通大学 | Human behavior intention recognition method based on target detection network and knowledge reasoning |
| CN114724078B (en) * | 2022-03-28 | 2023-06-23 | 西南交通大学 | Personnel Behavior Intent Recognition Method Based on Target Detection Network and Knowledge Reasoning |
| CN115862896A (en) * | 2023-02-13 | 2023-03-28 | 深圳市汇健智慧医疗有限公司 | Perioperative period-based doctor-patient cooperative management method and system |
| CN116468396A (en) * | 2023-04-14 | 2023-07-21 | 浪潮智慧科技有限公司 | Water conservancy service processing method, device and medium based on intent analysis model |
| CN116468396B (en) * | 2023-04-14 | 2023-11-03 | 浪潮智慧科技有限公司 | Water conservancy service processing method, device and medium based on intent analysis model |
| CN116842145A (en) * | 2023-04-20 | 2023-10-03 | 海信集团控股股份有限公司 | Domain identification method and device based on city question-answering system |
| CN116842145B (en) * | 2023-04-20 | 2024-02-27 | 海信集团控股股份有限公司 | Domain identification method and device based on city question-answering system |
| CN117034957A (en) * | 2023-06-30 | 2023-11-10 | 海信集团控股股份有限公司 | Semantic understanding method and device |
| CN117034957B (en) * | 2023-06-30 | 2024-05-31 | 海信集团控股股份有限公司 | Semantic understanding method and device integrating large models |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022041728A1 (en) | 2022-03-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112035637A (en) | Medical field intention recognition method, device, equipment and storage medium | |
| WO2022041730A1 (en) | Medical field intention recognition method, apparatus and device, and storage medium | |
| CN111538894B (en) | Query feedback method and device, computer equipment and storage medium | |
| CN112182252B (en) | Intelligent drug question answering method and equipment based on drug knowledge graph | |
| CN111813957A (en) | Medical diagnosis guiding method based on knowledge graph and readable storage medium | |
| CN113505243A (en) | Intelligent question-answering method and device based on medical knowledge graph | |
| US20220121824A1 (en) | Method for determining text similarity, method for obtaining semantic answer text, and question answering method | |
| CN109739964A (en) | Knowledge data providing method, device, electronic equipment and storage medium | |
| CN111341456B (en) | Method and device for generating diabetic foot knowledge graph and readable storage medium | |
| WO2013125286A1 (en) | Non-factoid question answering system and computer program | |
| CN113724882B (en) | Method, device, equipment and medium for constructing user portrait based on inquiry session | |
| CN110675944A (en) | Triage method and device, computer equipment and medium | |
| US11468070B2 (en) | Method and system for performing context-based search | |
| CN117056493A (en) | Large language model medical question-answering system based on medical record knowledge graph | |
| CN111723570A (en) | Medicine knowledge graph construction method and device and computer equipment | |
| JP2023514023A (en) | Question retrieval device, question retrieval method, device, and storage medium | |
| CN118013021B (en) | Medicine answering method, device, equipment and medium based on large language model | |
| CN111581972A (en) | Method, device, equipment and medium for identifying the corresponding relationship between symptoms and parts in text | |
| CN114153994A (en) | Medical insurance information question and answer method and device | |
| CN114004237A (en) | Intelligent question-answering system construction method based on bladder cancer knowledge graph | |
| CN112037904A (en) | Online diagnosis and treatment data processing method and device, computer equipment and storage medium | |
| CN117171329A (en) | Semantic analysis-based traditional Chinese medicine domain knowledge graph question-answering method | |
| CN114840657A (en) | API knowledge graph self-adaptive construction and intelligent question-answering method based on mixed mode | |
| WO2023185082A1 (en) | Training method and training device for language representation model | |
| Montenegro et al. | The HoPE model architecture: A novel approach to pregnancy information retrieval based on conversational agents |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201204 |
|
| RJ01 | Rejection of invention patent application after publication |