CN117974152A - Customer complaint data analysis method and device, storage medium and electronic equipment - Google Patents
Customer complaint data analysis method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN117974152A CN117974152A CN202410130833.3A CN202410130833A CN117974152A CN 117974152 A CN117974152 A CN 117974152A CN 202410130833 A CN202410130833 A CN 202410130833A CN 117974152 A CN117974152 A CN 117974152A
- Authority
- CN
- China
- Prior art keywords
- event
- complaint
- text
- customer
- word
- 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 59
- 238000007405 data analysis Methods 0.000 title claims abstract description 48
- 230000001364 causal effect Effects 0.000 claims abstract description 40
- 238000000605 extraction Methods 0.000 claims abstract description 32
- 239000013598 vector Substances 0.000 claims description 35
- 238000004590 computer program Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 abstract description 10
- 238000010276 construction Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012549 training Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
-
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a customer complaint data analysis method, a device, a storage medium and electronic equipment, wherein the method comprises the following steps: event extraction is carried out on the complaint text by acquiring the complaint text of customer complaints, event entities, event trigger words and types and attributes of the events in the complaint text are obtained, event relation extraction is carried out on the complaint text according to the event entities, the event trigger words and the types and attributes of the events, semantic and logic relations among the events are obtained, and a causal logic map of the complaint text is constructed according to the semantic and logic relations among the events. Therefore, the method and the system can automatically extract the event and the event relation from the text content of the customer complaints, construct the corresponding causal logic map, provide service support for customer service staff and improve the processing efficiency of the customer service staff on the customer complaints.
Description
Technical Field
The application relates to the technical field of data analysis, in particular to a customer complaint data analysis method, a customer complaint data analysis device, a storage medium and electronic equipment.
Background
Customer complaints are an important aspect in providing customer service to all current enterprises, reflecting customer dissatisfaction and expectations for products or services, and also being opportunities for improvement and promotion of the enterprises. However, the text content of customer complaints is often unstructured and contains information of various events, reasons, results, emotions and the like, and understanding and handling customer complaints is a time-consuming, complex and difficult task for customer service personnel. Especially when facing sudden, novel and complex customer complaints, customer service personnel often lack effective service support, and timely, accurate and satisfactory replies are difficult to give.
Disclosure of Invention
The embodiment of the application provides a customer complaint data analysis method, a device, a storage medium and electronic equipment, which can provide service support for customer service staff and improve the processing efficiency of the customer service staff on customer complaints.
In a first aspect, an embodiment of the present application provides a method for analyzing customer complaint data, including:
Acquiring complaint text of customer complaints;
Extracting the event from the complaint text to obtain an event entity, an event trigger word, and the type and attribute of the event in the complaint text;
Extracting event relation from the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain semantic and logic relation between the events;
And constructing a causal logic map of the complaint text according to the semantic and logical relations between the events.
In some embodiments, the extracting the event from the complaint text to obtain an event entity, an event trigger word, and a type and an attribute of the event in the complaint text includes:
Carrying out event boundary recognition on the complaint text, and determining event entities and event trigger words in the complaint text;
And classifying event types of the complaint text, and determining the types and the attributes of the events in the complaint text.
In some embodiments, the identifying the event boundary of the complaint text, determining the event entity and the event trigger word in the complaint text, includes:
Inputting the complaint text into a pre-trained language model to obtain word vectors corresponding to each word in the complaint text;
inputting the word vector into an event boundary recognition layer to obtain a word label corresponding to each word in the text;
and dividing the words in the complaint text into the event entity and the event trigger words according to the word labels.
In some embodiments, the classifying the event type of the complaint text, determining the type and the attribute of the event in the complaint text, includes:
Inputting the complaint text into a pre-trained language model to obtain word vectors and word position information corresponding to each word in the complaint text;
And inputting the word vector and the word position information into an event type classification layer to obtain the event type and the attribute.
In some embodiments, the extracting the event relation from the complaint text according to the event entity, the event trigger word, and the type and attribute of the event, to obtain the semantic and logical relation between the events, includes:
Constructing an event graph through the event entity, the event trigger word and the type and attribute of the event, wherein each node in the event graph represents the event;
Inputting the event map into a pre-trained map neural network to obtain a node vector corresponding to each node in the event map;
inputting the node vector into an event relation classification layer to obtain an edge label of each edge in the event graph;
And determining the semantic and logical relationship between the events in the event graph according to the edge labels.
In some embodiments, the acquiring complaint text of a customer complaint includes:
acquiring complaint data in a customer service system;
and carrying out data preprocessing on the complaint data to obtain the complaint text.
In some embodiments, after said constructing a causal logic map of said complaint text from semantic and logical relationships between said events, further comprising:
the cause and effect logic map is provided to a customer service system and the cause and effect logic map is displayed on the customer service system.
In some embodiments, after said constructing a causal logic map of said complaint text from semantic and logical relationships between said events, further comprising:
And storing the causal logic map into a preset map database.
In a second aspect, an embodiment of the present application further provides a customer complaint data analysis apparatus, including:
The acquiring unit is used for acquiring complaint texts of customer complaints;
the event extraction unit is used for extracting the event from the complaint text to obtain an event entity, an event trigger word, the type and the attribute of the event in the complaint text;
the event relation extraction unit is used for extracting the event relation of the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain the semantic and logic relation between the events;
A causal logic map construction unit, configured to construct a causal logic map of the complaint text according to the semantic and logical relationships between the events.
In a third aspect, embodiments of the present application also provide a computer-readable storage medium having stored thereon a computer program which, when run on a computer, causes the computer to perform a customer complaint data analysis method as provided by any of the embodiments of the present application.
In a fourth aspect, an embodiment of the present application further provides an electronic device, including a processor and a memory, where the memory has a computer program, and the processor is configured to execute the customer complaint data analysis method according to any embodiment of the present application by calling the computer program.
According to the technical scheme provided by the embodiment of the application, the complaint text of the customer complaint is obtained, event extraction is carried out on the complaint text to obtain event entities, event trigger words and types and attributes of the events in the complaint text, event relation extraction is carried out on the complaint text according to the event entities, the event trigger words and the types and attributes of the events to obtain semantic and logic relations among the events, and a causal logic map of the complaint text is constructed according to the semantic and logic relations among the events. Therefore, the method and the system can automatically extract the event and the event relation from the text content of the customer complaints, construct the corresponding causal logic map, provide service support for customer service staff and improve the processing efficiency of the customer service staff on the customer complaints.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a customer complaint data analysis method according to an embodiment of the present application.
Fig. 2 is an application scenario schematic diagram of a customer complaint data analysis method according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of a customer complaint data analysis device according to an embodiment of the present application.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by a person skilled in the art without any inventive effort, are intended to be within the scope of the present application based on the embodiments of the present application.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The embodiment of the application provides a customer complaint data analysis method, and an execution subject of the customer complaint data analysis method can be the customer complaint data analysis device provided by the embodiment of the application or electronic equipment integrated with the customer complaint data analysis device, wherein the customer complaint data analysis device can be realized in a hardware or software mode. The electronic device may be a smart phone, a tablet computer, a palm computer, a notebook computer, or a desktop computer.
Referring to fig. 1, fig. 1 is a flow chart of a method for analyzing customer complaint data according to an embodiment of the application. The specific flow of the customer complaint data analysis method provided by the embodiment of the application can be as follows:
s110, acquiring a complaint text of customer complaints.
Where complaint text generally refers to written complaint content presented by a consumer or customer to a related department or institution that is not satisfactory to the product, service, or institution itself.
The complaint text may be obtained by a customer service system. Wherein, customer service system refers to a system for managing customer service and support. Customer service systems typically include customer service support, complaint handling, work order management, online customer service, and like functions.
For example, the text content of customer complaints such as "I have ordered an air ticket on your website, and as a result, found to be false by the airport, your fraud, I ask for refunds and reimbursements-! ".
In some embodiments, the step of "obtaining complaint text of customer complaints" may include the steps of:
acquiring complaint data in a customer service system;
and carrying out data preprocessing on the complaint data to obtain the complaint text.
Wherein, the data preprocessing refers to the process of cleaning, converting and sorting the original data before data analysis or modeling. The purpose of data preprocessing is to improve the quality, accuracy and usability of data to facilitate subsequent data analysis, data mining or machine learning modeling.
For example, the format of complaint data may include various types such as audio format, image format, text format, and the like. The data preprocessing of the complaint data may be converting the format of the complaint data into a unified text format.
S120, extracting the event from the complaint text to obtain an event entity, an event trigger word, and the type and attribute of the event in the complaint text.
The event extraction means that event entities and event trigger words are identified from the text, and the types and attributes of the events. Event entities refer to entities involved in an event, such as people, things, places, etc.; an event trigger word refers to a word, such as a verb, noun, etc., that represents the occurrence of an event. The type and attribute of an event refer to the category and characteristics of the event, such as time, place, cause, result, etc. The purpose of event extraction is to convert event information in text into structured data, which provides a basis for subsequent event relation extraction and causal logic map construction.
In some embodiments, the step of "identifying an event boundary for the complaint text, determining an event entity and an event trigger word in the complaint text" may include the steps of:
Carrying out event boundary recognition on the complaint text, and determining event entities and event trigger words in the complaint text;
And classifying event types of the complaint text, and determining the types and the attributes of the events in the complaint text.
Wherein event boundary recognition is a sequence tagged task for determining whether each word in the text is an event entity or event trigger word, and their beginning and ending locations.
In some embodiments, the step of identifying an event boundary for the complaint text and determining an event entity and an event trigger word in the complaint text may include the steps of:
Inputting the complaint text into a pre-trained language model to obtain word vectors corresponding to each word in the complaint text;
inputting the word vector into an event boundary recognition layer to obtain a word label corresponding to each word in the text;
and dividing the words in the complaint text into the event entity and the event trigger words according to the word labels.
In this embodiment, the obtained complaint text is used as input of a pre-training language model to obtain word vectors (i.e., word vector representations) of each word in the complaint text, and then the word vectors of each word are transferred into an event boundary recognition layer to obtain labels of each word in the complaint text, such as B-event entity, I-event entity, B-event trigger word, I-event trigger word, etc. Then dividing the words in the complaint text into event entities and event trigger words according to the word labels, and recording the starting and ending positions of the event entities and the event trigger words.
The event extraction method adopted in the application is a method based on a pre-trained language model, and the method utilizes large-scale unlabeled text data to learn the semantic and grammar information of the text in a self-supervision mode, thereby improving the accuracy and the robustness of event extraction. The pre-trained language model fully utilizes the context information of the complaint text through a bidirectional attention mechanism to generate vector representation of each word in the complaint text. For example, the language model may include language models such as GPT-3, BERT, ERNIE 3.0, and T5.
In some embodiments, the step of "classifying the event type of the complaint text, determining the type and attribute of the event in the complaint text" may include the steps of:
the event type classification is a multi-classification task for judging the type and attribute of an event entity or event trigger word, such as time, place, reason, result and the like.
In some embodiments, the step of classifying the event type of the complaint text, determining the type and attribute of the event in the complaint text may include the steps of:
Inputting the complaint text into a pre-trained language model to obtain word vectors and word position information corresponding to each word in the complaint text;
And inputting the word vector and the word position information into an event type classification layer to obtain the event type and the attribute.
In this embodiment, the obtained complaint text is used as input of a pre-training language model to obtain word vectors and word position information of each word in the complaint text, and then the word vectors and the word position information are used as input to be transmitted into an event type classification layer to obtain types and attributes of the event, such as a main body, an object, a place, an article, a behavior, a judgment, a result, a reason and the like.
The input of the event type classification layer in this embodiment may reuse the word vector and word position information obtained by inputting the complaint text into the pre-trained language model.
S130, extracting the event relation from the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain the semantic and logic relation between the events.
The event relation extraction refers to identifying semantic and logical relations between events from texts, such as causal relations, conditional relations, sequential relations and the like. The purpose of event relation extraction is to convert event relations in the text into structured data, which provides a basis for subsequent causal logic map construction.
In some embodiments, the step of extracting the event relation from the complaint text according to the event entity, the event trigger word, and the type and attribute of the event to obtain the semantic and logical relation between the events may include the following steps:
Constructing an event graph through the event entity, the event trigger word and the type and attribute of the event, wherein each node in the event graph represents the event;
Inputting the event map into a pre-trained map neural network to obtain a node vector corresponding to each node in the event map;
inputting the node vector into an event relation classification layer to obtain an edge label of each edge in the event graph;
And determining the semantic and logical relationship between the events in the event graph according to the edge labels.
In this embodiment, an event entity and an event trigger word, and types and attributes of events are used as inputs to construct an event graph, where nodes represent events and edges represent potential relationships between events. And taking the event graph as input, transmitting the event graph into a pre-trained graph neural network to obtain vector representation of each node in the event graph, taking the vector representation as input, transmitting the vector representation into an event relation classification layer to obtain labels of each side in the event graph, such as causality, conditions, sequence, modification and the like, and screening the sides in the event graph into semantic and logic relations among the events according to the labels.
The event relation extraction method adopted by the application is a method based on a graph neural network, and by utilizing a graph structure, complex and various relations among events can be effectively represented, and the accuracy and the robustness of event relation extraction are improved. In the application, a GCN (Graph Convolutional Network) -based neural network can be used, and the network can propagate and aggregate information on a graph structure in a convolution mode to generate vector representation of each node in the graph. Then, the application adds a task specific layer for event relationship classification based on the GCN. Event relationship classification is a multi-classification task for determining the type of relationship between events, such as causal relationships, conditional relationships, sequential relationships, and the like. The application realizes the function of event relation extraction by fine-tuning the GCN on a small amount of marked data.
S140, constructing a causal logic map of the complaint text according to the semantic and logical relations between the events.
The causal logic map refers to a knowledge map which represents events and event relations in the form of a graph, and can reflect causal logic and logic relations among the events, so as to provide visual service support for customer service personnel.
In this embodiment, a causal logic graph is constructed using the semantic and logical relationships between events as inputs, where nodes represent events, edges represent relationships between events, and labels on edges represent the types of relationships.
The construction of the causal logic map in the application refers to the process of storing the results of event extraction and event relation extraction as the data of a graph structure and generating the causal logic map.
The causal logic map construction method adopted in the application is a method based on a map database, and the method can efficiently store and inquire the data of the map structure by using the map database, thereby improving the efficiency and the expandability of causal logic map construction.
In particular, the application is not limited by the order of execution of the steps described, as some of the steps may be performed in other orders or concurrently without conflict.
It can be seen from the foregoing that, according to the method for analyzing customer complaint data provided by the embodiment of the present application, event extraction is performed on a complaint text by obtaining the complaint text of a customer complaint, so as to obtain an event entity, an event trigger word, and a type and an attribute of an event in the complaint text, event relation extraction is performed on the complaint text according to the event entity, the event trigger word, and the type and the attribute of the event, so as to obtain a semantic and a logic relation between the events, and a causal logic map of the complaint text is constructed according to the semantic and the logic relation between the events. The application can automatically extract events and event relations from the text content of customer complaints, construct a corresponding causal logic map, and help customer service personnel quickly, accurately and comprehensively understand the problems and demands of customers and the reasons and results of the problems, thereby giving timely, reasonable and satisfactory replies to the customers so as to improve the processing efficiency of the customer complaints by the customer service personnel.
In addition, the customer complaint data analysis method provided by the embodiment of the application adopts an event extraction method based on a pre-trained language model, and the method utilizes large-scale unlabeled text data to learn the semantic and grammar information of the text in a self-supervision mode, so that the accuracy and the robustness of event extraction and the complexity and the diversity of event relation extraction are improved.
In some embodiments, after said constructing a causal logic map of said complaint text from semantic and logical relationships between said events, further comprising:
And storing the causal logic map into a preset map database.
In this embodiment, the causal logic map is stored in a preset map database, which can facilitate support for subsequent queries and analysis.
In some embodiments, after said constructing a causal logic map of said complaint text from semantic and logical relationships between said events, further comprising:
the cause and effect logic map is provided to a customer service system and the cause and effect logic map is displayed on the customer service system.
In this embodiment, the causal logic map may be returned as output to the relevant customer service system, so as to provide visual service support for customer service personnel.
Referring to fig. 2, fig. 2 is a schematic diagram of an application scenario of a customer complaint data analysis method according to an embodiment of the present application.
For example, a customer submits complaint data to a customer service system, the customer service system processes the complaint data through a customer complaint processing system after receiving the complaint data, a causal logic map corresponding to the complaint data is obtained, the customer complaint processing system returns the causal logic map to the customer service system, then customer service personnel conduct complaint processing through the returned causal logic map, and a complaint processing result is given to the customer.
In one embodiment, a customer complaint data analysis apparatus is also provided. Referring to fig. 3, fig. 3 is a schematic structural diagram of a customer complaint data analysis apparatus 200 according to an embodiment of the application. Wherein the customer complaint data analysis apparatus 200 is applied to an electronic device, the customer complaint data analysis apparatus 200 includes an acquisition unit 201, an event extraction unit 202, an event relation extraction unit 203, and a causal logic map construction unit 204, as follows:
an acquiring unit 201, configured to acquire a complaint text of a customer complaint;
An event extraction unit 202, configured to extract an event from the complaint text, so as to obtain an event entity, an event trigger word, and a type and an attribute of the event in the complaint text;
An event relation extraction unit 203, configured to extract an event relation from the complaint text according to the event entity, the event trigger word, and the type and attribute of the event, so as to obtain a semantic and a logic relation between the events;
A causal logic map construction unit 204 for constructing a causal logic map of the complaint text from the semantic and logical relationships between the events.
In some embodiments, the event extraction unit 202 may be configured to:
Carrying out event boundary recognition on the complaint text, and determining event entities and event trigger words in the complaint text;
And classifying event types of the complaint text, and determining the types and the attributes of the events in the complaint text.
In some embodiments, the event extraction unit 202 may be configured to:
Inputting the complaint text into a pre-trained language model to obtain word vectors corresponding to each word in the complaint text;
inputting the word vector into an event boundary recognition layer to obtain a word label corresponding to each word in the text;
and dividing the words in the complaint text into the event entity and the event trigger words according to the word labels.
In some embodiments, the event extraction unit 202 may be configured to:
Inputting the complaint text into a pre-trained language model to obtain word vectors and word position information corresponding to each word in the complaint text;
And inputting the word vector and the word position information into an event type classification layer to obtain the event type and the attribute.
In some embodiments, the event relationship extraction unit 203 may be configured to:
Constructing an event graph through the event entity, the event trigger word and the type and attribute of the event, wherein each node in the event graph represents the event;
Inputting the event map into a pre-trained map neural network to obtain a node vector corresponding to each node in the event map;
inputting the node vector into an event relation classification layer to obtain an edge label of each edge in the event graph;
And determining the semantic and logical relationship between the events in the event graph according to the edge labels.
In some embodiments, the acquisition unit 201 may be configured to:
acquiring complaint data in a customer service system;
and carrying out data preprocessing on the complaint data to obtain the complaint text.
In some embodiments, customer complaint data analysis apparatus 200 may further include a presentation unit that may be used to:
the cause and effect logic map is provided to a customer service system and the cause and effect logic map is displayed on the customer service system.
In some embodiments, customer complaint data analysis apparatus 200 may further include a storage unit that may be used to:
And storing the causal logic map into a preset map database.
It should be noted that, the customer complaint data analysis device provided by the embodiment of the present application belongs to the same concept as the customer complaint data analysis method in the above embodiment, and any method provided in the customer complaint data analysis method embodiment may be implemented by the customer complaint data analysis device, and the specific implementation process is detailed in the customer complaint data analysis method embodiment and will not be described herein.
In addition, in order to better implement the customer complaint data analysis method according to the embodiment of the present application, the present application further provides an electronic device based on the customer complaint data analysis method, referring to fig. 4, fig. 4 shows a schematic structural diagram of an electronic device 300 provided by the present application, and as shown in fig. 4, the electronic device 300 provided by the present application includes a processor 301 and a memory 302, where the processor 301 is configured to implement steps of the customer complaint data analysis method according to the above embodiment of the present application when executing a computer program stored in the memory 302, for example:
Acquiring complaint text of customer complaints;
Extracting the event from the complaint text to obtain an event entity, an event trigger word, and the type and attribute of the event in the complaint text;
Extracting event relation from the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain semantic and logic relation between the events;
And constructing a causal logic map of the complaint text according to the semantic and logical relations between the events.
By way of example, a computer program may be partitioned into one or more modules/units that are stored in memory 302 and executed by processor 301 to accomplish an embodiment of the application. One or more of the modules/units may be a series of computer program instruction segments capable of performing particular functions to describe the execution of the computer program in a computer device.
Electronic device 300 may include, but is not limited to, a processor 301, a memory 302. It will be appreciated by those skilled in the art that the illustration is merely an example of the electronic device 300 and is not limiting of the electronic device 300, and may include more or fewer components than shown, or may combine some of the components, or different components, e.g., the electronic device 300 may further include an input-output device, a network access device, a bus, etc., through which the processor 301, the memory 302, the input-output device, the network access device, etc., are connected.
The Processor 301 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), off-the-shelf Programmable gate array (Field-Programmable GATE ARRAY, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or any conventional processor or the like that is a control center of the electronic device 300 that interfaces and lines to various portions of the overall electronic device 300.
The memory 302 may be used to store computer programs and/or modules, and the processor 301 implements various functions of the computer device by running or executing the computer programs and/or modules stored in the memory 302 and invoking data stored in the memory 302. The memory 302 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data (such as audio data, video data, etc.) created according to the use of the electronic device 300, and the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart memory card (SMART MEDIA CARD, SMC), secure Digital (SD) card, flash memory card (FLASH CARD), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the customer complaint data analysis device, the electronic apparatus 300 and the corresponding units thereof described above may refer to the description of the customer complaint data analysis method in the above embodiment of the present application, and the description is omitted herein.
Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.
To this end, an embodiment of the present application provides a computer readable storage medium having stored therein a plurality of instructions capable of being loaded by a processor to perform the steps in the customer complaint data analysis method of the above embodiment of the present application, such as:
Acquiring complaint text of customer complaints;
Extracting the event from the complaint text to obtain an event entity, an event trigger word, and the type and attribute of the event in the complaint text;
Extracting event relation from the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain semantic and logic relation between the events;
And constructing a causal logic map of the complaint text according to the semantic and logical relations between the events.
For specific operations, reference may be made to the description of the method for analyzing customer complaint data in the above embodiments of the present application, and details thereof are not repeated herein.
Wherein the computer-readable storage medium may comprise: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like.
Because the instructions stored in the computer readable storage medium may execute the steps in the customer complaint data analysis method according to the above embodiment of the present application, the beneficial effects that can be achieved by the customer complaint data analysis method according to the above embodiment of the present application can be achieved, and detailed descriptions thereof will be omitted herein.
Furthermore, the terms "first," "second," and "third," and the like, herein, are used for distinguishing between different objects and not for describing a particular sequential order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or modules is not limited to the particular steps or modules listed and certain embodiments may include additional steps or modules not listed or inherent to such process, method, article, or apparatus.
The foregoing has described in detail a method, apparatus, electronic device and storage medium for analyzing customer complaint data, and specific examples have been used herein to illustrate the principles and embodiments of the present application, and the above description of the examples is only for aiding in understanding the method and core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, the present description should not be construed as limiting the present application.
Claims (10)
1. A method of analyzing customer complaint data, comprising:
Acquiring complaint text of customer complaints;
Extracting the event from the complaint text to obtain an event entity, an event trigger word, and the type and attribute of the event in the complaint text;
Extracting event relation from the complaint text according to the event entity, the event trigger word and the type and attribute of the event to obtain semantic and logic relation between the events;
And constructing a causal logic map of the complaint text according to the semantic and logical relations between the events.
2. The method for analyzing customer complaint data according to claim 1, wherein the event extraction is performed on the complaint text to obtain an event entity, an event trigger word, and a type and an attribute of an event in the complaint text, including:
Carrying out event boundary recognition on the complaint text, and determining event entities and event trigger words in the complaint text;
And classifying event types of the complaint text, and determining the types and the attributes of the events in the complaint text.
3. The method for analyzing customer complaint data according to claim 2, wherein the event boundary recognition is performed on the complaint text, and determining event entities and event trigger words in the complaint text includes:
Inputting the complaint text into a pre-trained language model to obtain word vectors corresponding to each word in the complaint text;
inputting the word vector into an event boundary recognition layer to obtain a word label corresponding to each word in the text;
and dividing the words in the complaint text into the event entity and the event trigger words according to the word labels.
4. A customer complaint data analysis method as claimed in claim 3 wherein said classifying the type of event in the complaint text, determining the type and nature of the event in the complaint text, comprises:
Inputting the complaint text into the pre-trained language model to obtain word vectors and word position information corresponding to each word in the complaint text;
And inputting the word vector and the word position information into an event type classification layer to obtain the event type and the attribute.
5. The method for analyzing customer complaint data according to claim 1, wherein the extracting the event relation from the complaint text according to the event entity, the event trigger word, and the type and attribute of the event to obtain the semantic and logical relation between the events includes:
Constructing an event graph through the event entity, the event trigger word and the type and attribute of the event, wherein each node in the event graph represents the event;
Inputting the event map into a pre-trained map neural network to obtain a node vector corresponding to each node in the event map;
inputting the node vector into an event relation classification layer to obtain an edge label of each edge in the event graph;
And determining the semantic and logical relationship between the events in the event graph according to the edge labels.
6. The method for analyzing customer complaint data according to claim 1, wherein the acquiring the complaint text of the customer complaint includes:
acquiring complaint data in a customer service system;
and carrying out data preprocessing on the complaint data to obtain the complaint text.
7. The customer complaint data analysis method of claim 1 further comprising, after said constructing a causal logic map of said complaint text from semantic and logical relationships between said events:
the cause and effect logic map is provided to a customer service system and the cause and effect logic map is displayed on the customer service system.
8. A customer complaint data analysis apparatus, characterized by comprising means for performing the customer complaint data analysis method as claimed in any one of claims 1 to 7.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when run on a computer, causes the computer to perform the customer complaint data analysis method as claimed in any one of claims 1 to 7.
10. An electronic device comprising a processor and a memory, the memory storing a computer program, wherein the processor is configured to perform the customer complaint data analysis method of any one of claims 1 to 7 by invoking the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410130833.3A CN117974152A (en) | 2024-01-30 | 2024-01-30 | Customer complaint data analysis method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410130833.3A CN117974152A (en) | 2024-01-30 | 2024-01-30 | Customer complaint data analysis method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117974152A true CN117974152A (en) | 2024-05-03 |
Family
ID=90850991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410130833.3A Pending CN117974152A (en) | 2024-01-30 | 2024-01-30 | Customer complaint data analysis method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117974152A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118643828A (en) * | 2024-08-15 | 2024-09-13 | 深圳市华傲数据技术有限公司 | Customer service appeal data processing method and device, electronic equipment and storage medium |
-
2024
- 2024-01-30 CN CN202410130833.3A patent/CN117974152A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118643828A (en) * | 2024-08-15 | 2024-09-13 | 深圳市华傲数据技术有限公司 | Customer service appeal data processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11663411B2 (en) | Ontology expansion using entity-association rules and abstract relations | |
| CN110168535B (en) | Information processing method and terminal, computer storage medium | |
| CN111191445B (en) | Advertisement text classification method and device | |
| US20160085855A1 (en) | Perspective data analysis and management | |
| CN109325121B (en) | Method and device for determining keywords of text | |
| US11354345B2 (en) | Clustering topics for data visualization | |
| CN117974152A (en) | Customer complaint data analysis method and device, storage medium and electronic equipment | |
| JP2022028881A (en) | Method of automatically generating advertisements, apparatus, device, and computer-readable storage medium | |
| US10042913B2 (en) | Perspective data analysis and management | |
| CN113407677B (en) | Method, apparatus, device and storage medium for evaluating consultation dialogue quality | |
| US10762089B2 (en) | Open ended question identification for investigations | |
| CN115099239A (en) | Resource identification method, device, equipment and storage medium | |
| CN113553431B (en) | User tag extraction method, device, equipment and medium | |
| CN118113852A (en) | Financial problem answering method, device, equipment, system, medium and product | |
| CN112989050A (en) | Table classification method, device, equipment and storage medium | |
| CN109146306B (en) | Enterprise management system | |
| US20210295036A1 (en) | Systematic language to enable natural language processing on technical diagrams | |
| CN113360672B (en) | Method, apparatus, device, medium and product for generating knowledge graph | |
| CN115017385A (en) | Article searching method, device, equipment and storage medium | |
| KR20230059364A (en) | Public opinion poll system using language model and method thereof | |
| US20210312223A1 (en) | Automated determination of textual overlap between classes for machine learning | |
| CN115062136A (en) | Event disambiguation method based on graph neural network and related equipment thereof | |
| CN114676694A (en) | Method, device, equipment, medium and program product for generating business model | |
| US11120204B2 (en) | Comment-based article augmentation | |
| CN113094471A (en) | Interactive data processing method and device |
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 |