CN105404618B - A kind of dialog text treating method and apparatus - Google Patents
A kind of dialog text treating method and apparatus Download PDFInfo
- Publication number
- CN105404618B CN105404618B CN201410470306.3A CN201410470306A CN105404618B CN 105404618 B CN105404618 B CN 105404618B CN 201410470306 A CN201410470306 A CN 201410470306A CN 105404618 B CN105404618 B CN 105404618B
- Authority
- CN
- China
- Prior art keywords
- deviation
- semantic
- conversation
- determining
- text
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 18
- 230000004044 response Effects 0.000 claims description 171
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000003672 processing method Methods 0.000 abstract description 10
- 230000008569 process Effects 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000006872 improvement 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
- 230000000750 progressive effect Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Machine Translation (AREA)
Abstract
The embodiment of the invention discloses a kind of dialog text processing method and processing device, method includes：Obtain the dialog text of dialogue；The current semantics irrelevance and history semanteme irrelevance of the dialogue are determined according to the dialog text；The current semantics irrelevance is used to indicate attendant in the dialogue for the answer matching degree of the proposed problem of user；The history semanteme irrelevance is used to indicate attendant and follows degree for the theme of the proposed problem of user in the dialogue；Service offset value of the attendant in the dialogue is determined according to the current semantics irrelevance and the history semanteme irrelevance.The embodiment of the present invention can carry out more accurate quantitative measurment to the service of attendant in dialogue.
Description
Technical Field
The present invention relates to the field of data processing, and in particular, to a method and an apparatus for processing dialog text data.
Background
In the field of providing voice or text service for users by service personnel, a large amount of voice and text service data are generated in the process of providing service for users by the service personnel. How to measure the service provided by service personnel for users quantitatively according to the generated voice and text service data is a problem to be solved.
At present, voice service data between service personnel and users are generally converted into corresponding text data, and the service of the service personnel is measured in the modes of keyword extraction, spot check and the like on the basis of the text data.
Disclosure of Invention
The embodiment of the invention provides a method and a device for processing conversation text data, which can be used for more accurately and quantitatively measuring the service of a service staff in a conversation.
In a first aspect, an embodiment of the present invention provides a method for processing dialog text data, including:
acquiring a dialog text of a dialog;
determining the current semantic deviation degree and the historical semantic deviation degree of the dialog according to the dialog text; the current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the questions proposed by the user in the dialogue; the historical semantic deviation degree is used for indicating the subject following degree of the service personnel for the questions posed by the user in the conversation;
and determining a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
Optionally, determining a current semantic deviation degree of the dialog according to the dialog text includes:
determining the theme deviation value of each response according to the user question text and the server reply text of each response in the conversation;
and calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
Optionally, determining a topic deviation value of each response according to the user question text and the service person response text of each response in the dialog comprises:
determining a semantic set subject to which a user question text of each response belongs and a semantic set subject to which a service personnel reply text belongs;
and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
Optionally, calculating a current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog, including:
and calculating the average value of the theme deviation values of all responses in the dialog, and taking the calculated average value as the current semantic deviation degree of the dialog.
Optionally, determining a historical semantic deviation degree of the dialog according to the text of the dialog includes:
determining user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses in the conversation, and calculating the user theme deviation values of the conversation according to the user theme deviation values of the two adjacent responses;
determining the service personnel theme deviation value of the two adjacent responses according to the service personnel answer texts of the two adjacent responses in the conversation, and calculating the service personnel theme deviation value of the conversation according to the service personnel theme deviation value of the two adjacent responses;
and calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
Optionally, determining the theme deviation value of the users who answer twice in the dialog according to the question texts of the users who answer twice in the dialog includes:
determining a semantic set subject to which a user question text of each response in the conversation belongs; determining deviation values between the semantic set topics to which the user question texts responded twice adjacently belong, and taking the determined deviation values as the deviation values of the user topics responded twice adjacently;
determining the subject deviation value of the service personnel who answer twice in the conversation according to the answer texts of the service personnel who answer twice in the conversation, wherein the method comprises the following steps:
determining a semantic set subject to which a response text of a service person responding each time in the conversation belongs; and determining the deviation value between the topics of the semantic set to which the two adjacent responded server answer texts belong, and taking the determined deviation value as the topic deviation value of the two adjacent responded server answers.
Optionally, calculating the user topic deviation value of the dialog according to the user topic deviation values of the two adjacent answers of the session includes:
calculating the sum of the user theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the user theme deviation value of the conversation;
calculating the service person theme deviation value of the conversation according to the service person theme deviation values of the two adjacent answers of the conversation, wherein the calculation comprises the following steps:
and calculating the sum of the service person theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the service person theme deviation value of the conversation.
Optionally, calculating the historical semantic deviation according to the user topic deviation value of the dialog and the service staff topic deviation value of the dialog includes:
and calculating a difference value between the user theme deviation value of the conversation and the service staff theme deviation value of the conversation, and taking a calculation result as the historical semantic deviation degree.
Optionally, determining a semantic set topic to which the user question text of each response belongs includes:
for the user question text which is responded each time, determining a keyword set of the user question text which is responded the time;
calculating semantic relevance between the keyword set and each semantic set subject;
determining a semantic set subject with the highest semantic relevance value with the keyword set as a semantic set subject to which a user questioning text of the response belongs;
determining the semantic set topic to which the service personnel reply text of each response belongs, comprising:
for the answer text of the service personnel who answer each time, determining a keyword set of the answer text of the service personnel who answer the time;
calculating semantic correlation between the keyword set of the service personnel reply text and each semantic set subject;
and determining a semantic set subject with the highest value of semantic relevance with the keyword set of the service personnel reply text as the semantic set subject to which the service personnel reply text of the reply belongs.
Optionally, determining a semantic set topic to which the user question text of each response belongs includes:
extracting a first keyword from the user question text which is responded each time, and determining the semantic set theme containing the first keyword as the semantic set theme to which the user question text which is responded each time belongs;
determining the semantic set topic to which the service personnel reply text of each response belongs, comprising:
and extracting a second keyword from the answer text of the service personnel for each answer, and determining the semantic set topic containing the second keyword as the semantic set topic to which the answer text of the service personnel for each answer belongs.
Optionally, determining a service offset value of a service person in the dialog according to the current semantic deviation and the historical semantic deviation includes:
calculating a service offset value for the service person in the session according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents a service offset value for a service person in the session; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
In a second aspect, an embodiment of the present invention provides a device for processing dialog text data, including:
an acquisition unit configured to acquire a dialog text of a dialog;
the first determining unit is used for determining the current semantic deviation degree and the historical semantic deviation degree of the dialog according to the dialog text; the current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the questions proposed by the user in the dialogue; the historical semantic deviation degree is used for indicating the subject following degree of the service personnel for the questions posed by the user in the conversation;
and the second determining unit is used for determining a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
Optionally, the first determining unit includes:
the first determining subunit is used for determining the theme deviation value of each response according to the user question text and the service staff response text of each response in the conversation;
and the first calculating subunit is used for calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
Optionally, the first determining subunit is specifically configured to:
determining a semantic set subject to which a user question text of each response belongs and a semantic set subject to which a service personnel reply text belongs;
and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
Optionally, the first calculating subunit is specifically configured to:
and calculating the average value of the theme deviation values of all responses in the dialog, and taking the calculated average value as the current semantic deviation degree of the dialog.
Optionally, the first determining unit includes:
the second determining subunit is used for determining the user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses in the conversation and calculating the user theme deviation values of the conversation according to the user theme deviation values of the two adjacent responses; determining the service personnel theme deviation value of the two adjacent responses according to the service personnel answer texts of the two adjacent responses in the conversation, and calculating the service personnel theme deviation value of the conversation according to the service personnel theme deviation value of the two adjacent responses;
and the second calculating subunit is used for calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
Optionally, the second determining subunit is specifically configured to:
determining a semantic set subject to which a user question text of each response in the conversation belongs; determining deviation values between the semantic set topics to which the user question texts responded twice adjacently belong, and taking the determined deviation values as the deviation values of the user topics responded twice adjacently;
determining a semantic set subject to which a response text of a service person responding each time in the conversation belongs; and determining the deviation value between the topics of the semantic set to which the two adjacent responded server answer texts belong, and taking the determined deviation value as the topic deviation value of the two adjacent responded server answers.
Optionally, the second determining subunit is specifically configured to:
calculating the sum of the user theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the user theme deviation value of the conversation;
and calculating the sum of the service person theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the service person theme deviation value of the conversation.
Optionally, the second calculating subunit is specifically configured to: and calculating a difference value between the user theme deviation value of the conversation and the service staff theme deviation value of the conversation, and taking a calculation result as the historical semantic deviation degree.
Optionally, the second determining unit is specifically configured to: calculating a service offset value for the service person in the session according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents a service offset value for a service person in the session; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
In the embodiment, a dialog text of a dialog is obtained, the current semantic deviation degree and the historical semantic deviation degree of the dialog are determined according to the dialog text, and the service deviation value of a service person in the dialog is determined according to the current semantic deviation degree and the historical semantic deviation degree, so that the service of the service person in the dialog is quantitatively embodied, and the quantitative measurement of the service person in the dialog is realized; in the measuring process, the service deviant of the service personnel in the conversation is determined according to the two aspects of the matching degree of the answer of the service personnel to the question presented by the user in the conversation and the subject following degree of the service personnel to the question presented by the user in the conversation, so that the measuring result can accurately represent the service of the service personnel in the conversation.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow chart of an embodiment of a dialog text data processing method according to the present invention;
FIG. 2 is a flow chart of another embodiment of a dialog text data processing method according to the invention;
FIG. 3A is a flowchart of a method for implementing step 202 of the present invention;
FIG. 3B is a diagram illustrating the relationship between the user question text and the service person answer text in the dialog of the present invention and the subject of the semantic collection;
FIG. 3C is a flowchart of a method for implementing step 204 of the present invention;
FIG. 3D is a flowchart of an implementation of step 205 of the present invention;
FIG. 4 is a block diagram of an exemplary dialog text data processing device according to the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
The embodiment of the invention can be applied to any intelligent terminal, and specifically, the terminal can be a personal computer, a server and the like.
The method for processing the dialogue text data can be applied to evaluation of the service quality of the service personnel in the dialogue and used as basic data for evaluation of the service quality of the service personnel. For example, the dialog text data processing method of the present invention may be used to process dialog text data of each dialog in which a certain service person participates, obtain a processing result of each dialog, and evaluate and analyze the service quality of the service person; or, the dialogue text data processing method of the invention can be used for processing the dialogue text data of a large number of dialogues to obtain the processing result of each dialogue, thereby evaluating and analyzing the service quality of the service personnel as a whole and obtaining quality evaluation reports of different views and visual angles.
Or, the dialogue text data processing method can also be applied to dialogue real-time control as basic data of the dialogue real-time control. For example, the dialog text data processing method of the present invention may be executed in real time during a dialog, measure the service of the service person, and control the service person to continue the dialog or replace the service person of the dialog according to the real-time processing result.
Of course, the method for processing dialog text data of the present invention can also be applied to other application scenarios, which are not described herein again.
Referring to fig. 1, a schematic diagram of an embodiment of a dialog text data processing method according to the present invention includes:
step 101: dialog text for a dialog is obtained.
The dialog in the embodiment of the present invention may be a voice or a text, and when the dialog is a voice, the dialog may be converted from a voice to a text before the embodiment of the present invention is executed.
Step 102: and determining the current semantic deviation degree and the historical semantic deviation degree of the dialog according to the dialog text.
The current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the question proposed by the user in each answer of the conversation; the historical semantic deviation is used to indicate the subject following degree of the service personnel to the questions posed by the user in the dialog.
The larger the value of the current semantic deviation degree is, the lower the answer matching degree of the service personnel to the question proposed by the user in each answer of the conversation is, and the higher the answer matching degree is otherwise;
the larger the value of the historical semantic deviation degree is, the lower the subject following degree of the service personnel for the problem posed by the user in the conversation is, and the higher the subject following degree is.
Step 103: and determining a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
When the embodiment is applied to service quality evaluation of service personnel in a conversation, the service offset value of the service personnel in the conversation can quantitatively reflect the service quality of the service personnel in the conversation, and the smaller the obtained service offset value is, the higher the service quality of the service personnel in the conversation is, the larger the obtained service offset value is, and the lower the service quality of the service personnel in the conversation is.
In the embodiment, a dialog text of a dialog is obtained, the current semantic deviation degree and the historical semantic deviation degree of the dialog are determined according to the dialog text, and the service deviation value of a service worker in the dialog is determined according to the current semantic deviation degree and the historical semantic deviation degree, so that the service deviation value of the service worker in the dialog can be determined through the dialog text, the service of the service worker in the dialog is quantitatively embodied, and the quantitative measurement of the service worker in the dialog is realized; in the measuring process, the service deviation value of the service personnel in the conversation is determined from two aspects of the answer matching degree of the service personnel to the question proposed by the user in each answer of the conversation and the topic following degree of the service personnel to the question proposed by the user in the conversation, so that the measuring result can accurately represent the service of the service personnel in the conversation.
Furthermore, when the embodiment is applied to the evaluation of the service quality of the service personnel in the conversation, the service quality of the service personnel in the conversation can be quantitatively measured accurately.
Referring to fig. 2, a schematic diagram of another embodiment of the dialog text data processing method of the present invention includes:
step 201: dialog text for a dialog is obtained.
Each dialog includes a plurality of dialogs of the user and the service personnel, the question of the user and the response of the service personnel to the question constitute a response in the dialog, therefore, the dialog text of the dialog is composed of a plurality of user uttered texts and the text uttered by the service personnel, and the dialog in the embodiment of the invention is the dialog under the scene that the service personnel provides the service for the user, so the dialog generally includes: response text of a plurality of responses; the response text of each response includes: text of a user question (hereinafter, referred to as user question text) and text of a response of a service person to the question (hereinafter, referred to as service person response text).
When the service personnel provides service for the user, the conversation can be initiated by the user, the service personnel correspondingly answers according to the user questions, the first section of text of the conversation text is the user question text, and when the conversation text is divided into the response texts, each section of user question text and the response text of the service personnel following the user question text are sequentially divided into the response text of one response.
For example, assume that the dialog text is:
the user: you good!
Service personnel: your good!
The user: asking for what is the price of the item?
Service personnel: the price of this commodity is 50 yuan.
This dialog text includes: the response texts of 2 responses are "user: you good! Service personnel: your good! "," user: asking for what is the price of the item? Service personnel: the price of this commodity is 50 yuan. ".
Wherein, the answer text of the 1 st answer comprises: the user asks the text "user: you good! "and attendant reply text" attendant: your good! ", the response text of the 2 nd response includes: the user asks the text "user: asking for what is the price of the item? "and attendant reply text" attendant: the price of this commodity is 50 yuan. ".
When the service personnel provides service for the user, the conversation can also be initiated by the service personnel, the first section of text of the conversation text is the text spoken by the service personnel, and then the user question text and the service personnel reply text are sequentially carried out, at this time, when the conversation text is divided into the reply texts, the first section of text in the conversation text can be removed, and each section of user question text and the service personnel reply text which follows the user question text are sequentially divided into the reply texts which are responded for one time from the first section of user question text.
Step 202: and determining the theme deviation value of each response according to the user question text and the service person response text of each response in the conversation.
Wherein the topic deviation value of each response is used for indicating the response matching degree of the service personnel to the question proposed by the user in each response. The smaller the theme deviation value of a certain response, the higher the matching degree of the response of the service person to the question posed by the user in the response, and conversely, the lower the matching degree of the response of the service person to the question posed by the user in the response.
As shown in fig. 3A, the present step may include:
step 301: for each response in the dialog, determining the semantic set subject to which the user question text of each response belongs and the semantic set subject to which the service personnel answer text belongs.
In a first possible implementation manner, determining a semantic set topic to which the user question text of each response belongs may include:
for the user question text which is responded each time, determining a keyword set of the user question text;
calculating semantic relevance between the keyword set and each semantic set subject;
and determining a semantic set subject with the highest semantic relevance value with the keyword set as the semantic set subject to which the user question text of the response belongs.
When determining the keyword set of the user question text, the keyword set of the user question text can be obtained by a method of carrying out word division on the user question text and removing auxiliary words such as conjunctions and adverbs from the divided words. The specific removal of which auxiliary terms can be specifically set and defined in practical application, and the embodiment of the present invention is not limited. For example, suppose the user question text is "ask how to pay for the item? "then, the user question text may first be word divided into: asking, how, payment, the asking, the commodity and the expense, wherein the asking, the paying and the expense can be eliminated as auxiliary words, and the rest asking, the paying and the commodity and the expense form a keyword set of a user asking text.
The semantic relevance between the keyword set and each semantic set topic can be calculated through a relevant algorithm of the semantic relevance, and the specific algorithm is not limited in the embodiment of the invention. Because the algorithm implementation methods of semantic relevance are various, the semantic relevance is not listed one by one.
If two or more semantic set subjects are the same as and the highest in semantic correlation value with the keyword set, one semantic set subject can be determined from the semantic set subjects as the semantic set subject to which the user question text of the response belongs, or the priority of the semantic set subject can be preset, and the semantic set subject with the highest priority is determined as the semantic set subject to which the user question text of the response belongs, and the like.
The step of determining the semantic set topic to which the reply text of the service person for each reply belongs may include:
for the answer text of the service personnel answering each time, determining a keyword set of the answer text of the service personnel;
calculating semantic relevance between the keyword set and each semantic set subject;
and determining the semantic set subject with the highest value of semantic relevance with the keyword set as the semantic set subject of the answer service personnel answer text.
The specific implementation of the semantic set topic to which the service person answer text for each answer belongs may be determined by referring to the above-mentioned related description in the specific implementation of the semantic set topic to which the user question text for each answer belongs, and details are not repeated here.
In a second possible implementation manner, the step may include: extracting a first keyword from the user question text which is responded each time, determining a semantic set subject to which the user question text belongs according to the extracted first keyword, extracting a second keyword from the server answer text which is responded each time, and determining the semantic set subject to which the server answer text belongs according to the extracted second keyword.
The first and second keywords are only used for distinguishing the keywords are extracted from the user question text or the service personnel answer text, and have no other special meanings.
The keywords in the user question text or the service personnel reply text can be extracted by a related keyword extraction method, and the embodiment of the invention is not limited by a specific implementation method.
In practical application, business words possibly related to a service worker in a service process can be divided into at least two semantic set topics according to semantics, each semantic set topic comprises at least one word, and the words in different semantic set topics are different.
Preferably, the words set in the semantic collection theme are correlated with the keywords extracted in the keyword extraction, and particularly, the keywords to be extracted in the keyword extraction method are preferably words contained in the semantic collection theme.
For example, assuming that the preset semantic collection topics are shown in table 1 below, when extracting keywords from the user question text or the service person reply text, words included in each semantic collection topic, such as the word A, B, C, etc., may be extracted accordingly.
TABLE 1
| Semantic Collection topic 1 | Word A, word B, word C |
| Semantic Collection topic 2 | Word D, word E, word F |
| … | … |
| Semantic Collection topic n | Word X, word Y, word Z |
Correspondingly, when the semantic set subject to which the user question text belongs is determined according to the extracted first keyword, the semantic set subject containing the first keyword can be determined as the semantic set subject to which the user question text belongs.
For example, if the extracted first keyword is a word Y, the semantic set topic to which the user question text belongs is a semantic set topic n.
Similarly, when the semantic set topic to which the service person reply text belongs is determined according to the extracted second keyword, the semantic set topic containing the second keyword may be determined as the semantic set topic to which the service person reply text belongs.
For example, if the extracted second keyword is a word a, the semantic set topic to which the service person reply text belongs is semantic set topic 1.
Step 302: and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
In practical application, deviation values between semantic set topics may be preset, generally, deviation values between the same semantic set topics are 0, deviation values between other semantic set topics may be set according to semantic similarity of words included in the semantic set topics, the closer the semantics are, the smaller the deviation values between the semantic set topics are, and conversely, the larger the deviation values between the semantic set topics are, and a specific numerical value may be autonomously set in practical application, for example, as shown in table 2 below.
TABLE 2
| Semantic Collection topic 1 | Semantic Collection topic 2 | Semantic Collection topic x | Semantic Collection topic n | |
| Semantic Collection topic 1 | 0 | 3 | 5 | 4 |
| Semantic Collection topic 2 | 3 | 0 | 7 | 8 |
| Semantic Collection topic x | 5 | 7 | 0 | 5 |
| Semantic Collection topic n | 4 | 8 | 5 | 0 |
Correspondingly, in the step, the deviation value between the semantic set topic to which the user question text belongs and the semantic set topic to which the service personnel answer text belongs can be obtained by searching the deviation value between the preset semantic set topics. For example, assuming that the semantic set topic to which the user question text of a certain response belongs is the semantic set topic 1, and the semantic set topic to which the service person response text belongs is the semantic set topic n, it can be obtained that the deviation value between the semantic set topic to which the user question text of the response belongs and the semantic set topic to which the service person response text belongs is 4, that is, the deviation value of the topic of the response is 4.
In this step, the answer matching degree of the service staff to the user question in each answer can be obtained respectively by determining the topic deviation value of each answer in the conversation, and when the user question text and the semantic set topic to which the service staff answer text belongs in each answer are the same semantic set topic, the answer matching degree of the service staff to the user question is the highest.
For example, referring to fig. 3B, if the topics of the semantic sets to which the user question texts and the service person answer texts of the first to third responses in the dialog belong are the same, the topic deviation values of the first to third responses in the example combining table 2 are 0, the matching degree of the service person's answer to the user question is high, the topic deviation values of the user question texts and the service person answer texts of the fourth and fifth responses in the example combining table 2 are different, the topic deviation values of the fourth and fifth responses in the example combining table 2 are both greater than 0, and the matching degree of the service person's answer to the user question is relatively low.
Step 203: and calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
In this step, the average value of the topic deviation values of all responses in the dialog can be calculated and used as the current semantic deviation degree of the dialog.
Step 204: and determining the user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses of the conversation, and calculating the user theme deviation values of the conversation according to the user theme deviation values of the two adjacent responses.
Referring to fig. 3C, determining the user topic deviation values of two consecutive responses of the dialog based on the user question text of the two consecutive responses may include:
step 311: for each response in the dialog, determining the semantic set subject to which the user questioning text of each response belongs.
The implementation of this step may refer to the related description in step 301, which is not described herein again. In addition, when step 301 is executed before step 311, in this step 311, the execution result of step 301 may also be directly obtained, so as to obtain the semantic collection topic to which the user question text of each response belongs.
Step 312: and determining the deviation value between the semantic set topics to which the user question texts of the two adjacent responses belong in the conversation as the user topic deviation value of the two adjacent responses.
When the user topic deviation value of the dialog is calculated according to the user topic deviation values of the two adjacent responses, the sum of the user topic deviation values of the two adjacent responses of the dialog can be calculated as the user topic deviation value of the dialog.
Step 205: and determining the subject deviation value of the service personnel in the two adjacent responses according to the response text of the service personnel in the two adjacent responses of the conversation, and calculating the subject deviation value of the service personnel in the conversation according to the subject deviation value of the service personnel in the two adjacent responses.
Referring to fig. 3D, determining the server topic deviation values of two consecutive answers based on the server answer text of two consecutive answers to the conversation may include:
step 321: for each response in the dialog, the semantic set topic to which the response text of the attendant for each response belongs is determined.
The implementation of this step may refer to the related description in step 301, which is not described herein again. In addition, when step 301 is executed before step 311, in this step 311, the execution result of step 301 may also be directly obtained, so as to obtain the semantic collection topic to which the user question text of each response belongs.
Step 322: and determining the deviation value between the topics of the semantic set to which the service personnel answering twice in the conversation belong as the topic deviation value of the service personnel answering twice in the conversation.
When the user theme deviation value of the dialog is calculated according to the service person theme deviation values of two adjacent responses, the sum of the service person theme deviation values of two adjacent responses of the dialog can be calculated to serve as the user theme deviation value of the dialog.
Step 206: and calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
In this step, the difference between the user theme deviation value and the service staff theme deviation value may be calculated as the historical semantic deviation.
Wherein, the execution sequence between steps 202-203 and steps 204-206 is not limited; the execution order between step 204 and step 205 is not limited.
If step 202, step 204 and step 205 are implemented by fig. 3A, fig. 3C and fig. 3D, respectively, and step 301, step 311 and step 321 all need to determine the semantic set topic to which the user question text belongs or the semantic set topic to which the service person reply text belongs in each response, therefore, in order to improve the processing efficiency and reduce the data processing amount, the later executed step can directly obtain the execution result of the earlier executed step when determining the semantic set topic to which the user question text belongs or the semantic set topic to which the service person reply text belongs in each response.
The historical dialog semantic deviation can be used for judging the following situation of the fluctuation of the service personnel to the user-posed problem. The number of questions asked by the user, the accurate description of the questions, the topic conversion of the questions asked by the user can be finally reflected on the topic deviation value of the user in the conversation through the topic deviation values of the user answered twice adjacently, and the topic deviation value of the service staff who answers the questions asked by the service staff can be finally reflected on the topic deviation value of the service staff in the conversation through the topic deviation values of the service staff who answers twice adjacently, according to whether the topic conversion of the user questions is correspondingly performed, and the like. The historical semantic deviation obtained by subtracting the topic deviation value of the user and the topic deviation value of the service staff in the conversation indicates the topic following degree of the service staff for the user to ask a question in the whole conversation process, if the historical semantic deviation is positive, the service staff does not answer the user question in the rhythm of the change of the user question, and if the historical semantic deviation is negative, the service staff grasps the essence of the user question and can quickly answer the user question.
Step 207: and calculating a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
In this step, the service deviation value of the service staff in the session can be calculated according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents the service offset value of the service personnel in the conversation; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
The specific value of n can be set autonomously in the actual application, and in a possible implementation, different values of n can be set according to the number of responses included in the dialog, see table 3. Generally, the larger the number of responses, the smaller the value of the corresponding pi value. For example, in Table 3, Π 1 > Π 2 > Π 3 > Π 4 > Π 5 > Π 6.
TABLE 3
| Number of responses | 1-5 | 6-10 | 11-15 | 16-20 | 21-30 | Over 31 |
| Π | Π1 | Π2 | Π3 | Π4 | Π5 | Π6 |
In the embodiment, the service deviation value of the service personnel in the conversation can be determined through the conversation text, so that the service of the service personnel in the conversation is quantitatively embodied, and the quantitative measurement of the service personnel in the conversation is realized; in the measuring process, the service deviation value of the service personnel in the conversation is determined from two aspects of the answer matching degree of the service personnel to the question proposed by the user in each answer of the conversation and the topic following degree of the service personnel to the question proposed by the user in the conversation, so that the measuring result can accurately represent the service of the service personnel in the conversation.
Furthermore, when the embodiment is applied to the evaluation of the service quality of the service personnel in the conversation, the service quality of the service personnel in the conversation can be quantitatively measured accurately.
In correspondence with the above method, an embodiment of the present invention further provides a dialog text data processing apparatus, and as shown in fig. 4, the apparatus 400 includes: an acquisition unit 410, a first determination unit 420, and a second determination unit 430; wherein,
an acquisition unit 410 for acquiring a dialog text of a dialog;
a first determining unit 420, configured to determine a current semantic deviation degree and a historical semantic deviation degree of the dialog according to the dialog text; the current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the questions proposed by the user in the dialogue; the historical semantic deviation degree is used for indicating the subject following degree of the service personnel for the questions posed by the user in the conversation;
a second determining unit 430, configured to determine a service offset value of a service person in the dialog according to the current semantic deviation and the historical semantic deviation.
Optionally, the first determining unit 420 may include:
the first determining subunit is used for determining the theme deviation value of each response according to the user question text and the service staff response text of each response in the conversation;
and the first calculating subunit is used for calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
Optionally, the first determining subunit may be specifically configured to:
determining a semantic set subject to which a user question text of each response belongs and a semantic set subject to which a service personnel reply text belongs;
and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
Optionally, the first determining subunit may be specifically configured to:
for the user question text which is responded each time, determining a keyword set of the user question text which is responded the time;
calculating semantic relevance between the keyword set and each semantic set subject;
determining a semantic set subject with the highest semantic relevance value with the keyword set as a semantic set subject to which a user questioning text of the response belongs;
for the answer text of the service personnel who answer each time, determining a keyword set of the answer text of the service personnel who answer the time;
calculating semantic correlation between the keyword set of the service personnel reply text and each semantic set subject;
and determining a semantic set subject with the highest value of semantic relevance with the keyword set of the service personnel reply text as the semantic set subject to which the service personnel reply text of the reply belongs.
Optionally, the first computing subunit may be specifically configured to:
and calculating the average value of the theme deviation values of all responses in the dialog, and taking the calculated average value as the current semantic deviation degree of the dialog.
Optionally, the first determining unit 420 may include:
the second determining subunit is used for determining the user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses in the conversation and calculating the user theme deviation values of the conversation according to the user theme deviation values of the two adjacent responses; determining the service personnel theme deviation value of the two adjacent responses according to the service personnel answer texts of the two adjacent responses in the conversation, and calculating the service personnel theme deviation value of the conversation according to the service personnel theme deviation value of the two adjacent responses;
and the second calculating subunit is used for calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
Optionally, the second determining subunit may be specifically configured to:
determining a semantic set subject to which a user question text of each response in the conversation belongs; determining deviation values between the semantic set topics to which the user question texts responded twice adjacently belong, and taking the determined deviation values as the deviation values of the user topics responded twice adjacently;
determining a semantic set subject to which a response text of a service person responding each time in the conversation belongs; and determining the deviation value between the topics of the semantic set to which the two adjacent responded server answer texts belong, and taking the determined deviation value as the topic deviation value of the two adjacent responded server answers.
Optionally, the second determining subunit may be specifically configured to:
for the user question text which is responded each time, determining a keyword set of the user question text which is responded the time;
calculating semantic relevance between the keyword set and each semantic set subject;
determining a semantic set subject with the highest semantic relevance value with the keyword set as a semantic set subject to which a user questioning text of the response belongs;
for the answer text of the service personnel who answer each time, determining a keyword set of the answer text of the service personnel who answer the time;
calculating semantic correlation between the keyword set of the service personnel reply text and each semantic set subject;
and determining a semantic set subject with the highest value of semantic relevance with the keyword set of the service personnel reply text as the semantic set subject to which the service personnel reply text of the reply belongs.
Optionally, the second determining subunit may be specifically configured to:
calculating the sum of the user theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the user theme deviation value of the conversation;
and calculating the sum of the service person theme deviation values of the two adjacent responses of the conversation, and taking the calculation result as the service person theme deviation value of the conversation.
Optionally, the second calculating subunit may be specifically configured to: and calculating a difference value between the user theme deviation value of the conversation and the service staff theme deviation value of the conversation, and taking a calculation result as the historical semantic deviation degree.
Optionally, the second determining unit 430 may specifically be configured to: calculating a service offset value for the service person in the session according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents a service offset value for a service person in the session; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
In the embodiment, a dialog text of a dialog is obtained, the current semantic deviation degree and the historical semantic deviation degree of the dialog are determined according to the dialog text, and the service deviation value of a service worker in the dialog is determined according to the current semantic deviation degree and the historical semantic deviation degree, so that the service deviation value of the service worker in the dialog can be determined through the dialog text, the service of the service worker in the dialog is quantitatively embodied, and the quantitative measurement of the service worker in the dialog is realized; in the measuring process, the service deviation value of the service personnel in the conversation is determined from two aspects of the answer matching degree of the service personnel to the question proposed by the user in each answer of the conversation and the topic following degree of the service personnel to the question proposed by the user in the conversation, so that the measuring result can accurately represent the service of the service personnel in the conversation.
Furthermore, when the embodiment is applied to the evaluation of the service quality of the service personnel in the conversation, the service quality of the service personnel in the conversation can be quantitatively measured accurately.
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above-described embodiments of the present invention do not limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (20)
1. A method for processing dialog text data, comprising:
acquiring a dialog text of a dialog;
determining the current semantic deviation degree and the historical semantic deviation degree of the dialog according to the dialog text; the current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the questions proposed by the user in the dialogue; the historical semantic deviation degree is used for indicating the subject following degree of the service personnel for the questions posed by the user in the conversation;
and determining a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
2. The method of claim 1, wherein determining the current semantic deviation of the dialog from the dialog text comprises:
determining the theme deviation value of each response according to the user question text and the server reply text of each response in the conversation;
and calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
3. The method of claim 2, wherein determining a topic deviation value for each response from the user question text and the attendant answer text for each response in the dialog comprises:
determining a semantic set subject to which a user question text of each response belongs and a semantic set subject to which a service personnel reply text belongs;
and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
4. The method of claim 2, wherein calculating the current semantic deviation of the dialog based on the topic deviation values of all of the secondary responses in the dialog comprises:
and calculating the average value of the theme deviation values of all responses in the dialog, and taking the calculated average value as the current semantic deviation degree of the dialog.
5. The method of claim 1, wherein determining the historical semantic deviation of the dialog from the text of the dialog comprises:
determining user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses in the conversation, and calculating the user theme deviation values of the conversation according to the user theme deviation values of all the two adjacent responses in the conversation;
determining the subject deviation values of the service personnel of the two adjacent responses according to the answer texts of the service personnel of the two adjacent responses in the conversation, and calculating the subject deviation values of the service personnel of the conversation according to the subject deviation values of all the service personnel of the two adjacent responses in the conversation;
and calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
6. The method of claim 5, wherein determining the user topic deviation value of two adjacent responses according to the user question text of two adjacent responses in the dialog comprises:
determining a semantic set subject to which a user question text of each response in the conversation belongs; determining deviation values between the semantic set topics to which the user question texts responded twice adjacently belong, and taking the determined deviation values as the deviation values of the user topics responded twice adjacently;
determining the subject deviation value of the service personnel who answer twice in the conversation according to the answer texts of the service personnel who answer twice in the conversation, wherein the method comprises the following steps:
determining a semantic set subject to which a response text of a service person responding each time in the conversation belongs; and determining the deviation value between the topics of the semantic set to which the two adjacent responded server answer texts belong, and taking the determined deviation value as the topic deviation value of the two adjacent responded server answers.
7. The method of claim 5, wherein calculating the user topic deviation value of the dialog based on the user topic deviation values of two adjacent responses in the dialog comprises:
calculating the sum of the user theme deviation values of two adjacent responses in the conversation, and taking the calculation result as the user theme deviation value of the conversation;
calculating the service person theme deviation value of the conversation according to the service person theme deviation values of two adjacent answers in the conversation, wherein the calculation comprises the following steps:
and calculating the sum of the service personnel theme deviation values of two adjacent responses in the conversation, and taking the calculation result as the service personnel theme deviation value of the conversation.
8. The method of claim 5, wherein calculating the historical semantic deviation from the user topic deviation value of the conversation and the attendant topic deviation value of the conversation comprises:
and calculating a difference value between the user theme deviation value of the conversation and the service staff theme deviation value of the conversation, and taking a calculation result as the historical semantic deviation degree.
9. The method of claim 3 or 6, wherein determining the semantic collection topic to which the user question text of each response belongs comprises:
for the user question text which is responded each time, determining a keyword set of the user question text which is responded the time;
calculating semantic relevance between the keyword set and each semantic set subject;
determining a semantic set subject with the highest semantic relevance value with the keyword set as a semantic set subject to which a user questioning text of the response belongs;
determining the semantic set topic to which the service personnel reply text of each response belongs, comprising:
for the answer text of the service personnel who answer each time, determining a keyword set of the answer text of the service personnel who answer the time;
calculating semantic correlation between the keyword set of the service personnel reply text and each semantic set subject;
and determining a semantic set subject with the highest value of semantic relevance with the keyword set of the service personnel reply text as the semantic set subject to which the service personnel reply text of the reply belongs.
10. The method of claim 3 or 6, wherein determining the semantic collection topic to which the user question text of each response belongs comprises:
extracting first keywords from the user question texts which are responded each time, and determining a semantic set theme containing the first keywords as a semantic set theme to which the user question texts which are responded each time belong;
determining the semantic set topic to which the service personnel reply text of each response belongs, comprising:
and extracting a second keyword from the answer text of the service personnel for each answer, and determining the semantic set topic containing the second keyword as the semantic set topic to which the answer text of the service personnel for each answer belongs.
11. The method of any one of claims 1 to 8, wherein determining a service offset value for a service person in the conversation based on the current semantic deviation and the historical semantic deviations comprises:
calculating a service offset value for the service person in the session according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents a service offset value for a service person in the session; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
12. A dialogue text data processing apparatus, comprising:
an acquisition unit configured to acquire a dialog text of a dialog;
the first determining unit is used for determining the current semantic deviation degree and the historical semantic deviation degree of the dialog according to the dialog text; the current semantic deviation degree is used for indicating the answer matching degree of the service personnel to the questions proposed by the user in the dialogue; the historical semantic deviation degree is used for indicating the subject following degree of the service personnel for the questions posed by the user in the conversation;
and the second determining unit is used for determining a service deviation value of the service personnel in the conversation according to the current semantic deviation degree and the historical semantic deviation degree.
13. The apparatus of claim 12, wherein the first determining unit comprises:
the first determining subunit is used for determining the theme deviation value of each response according to the user question text and the service staff response text of each response in the conversation;
and the first calculating subunit is used for calculating the current semantic deviation degree of the dialog according to the topic deviation values of all responses in the dialog.
14. The apparatus according to claim 13, wherein the first determining subunit is specifically configured to:
determining a semantic set subject to which a user question text of each response belongs and a semantic set subject to which a service personnel reply text belongs;
and determining a deviation value between the semantic set subject to which the user question text belongs and the semantic set subject to which the service personnel answer text belongs for each answer, and taking the determined deviation value as the subject deviation value for each answer.
15. The apparatus according to claim 13, wherein the first computing subunit is specifically configured to:
and calculating the average value of the theme deviation values of all responses in the dialog, and taking the calculated average value as the current semantic deviation degree of the dialog.
16. The apparatus of claim 12, wherein the first determining unit comprises:
the second determining subunit is used for determining the user theme deviation values of the two adjacent responses according to the user question texts of the two adjacent responses in the conversation and calculating the user theme deviation values of the conversation according to the user theme deviation values of all the two adjacent responses in the conversation; determining the subject deviation values of the service personnel of the two adjacent responses according to the answer texts of the service personnel of the two adjacent responses in the conversation, and calculating the subject deviation values of the service personnel of the conversation according to the subject deviation values of all the service personnel of the two adjacent responses in the conversation;
and the second calculating subunit is used for calculating the historical semantic deviation degree according to the user theme deviation value of the conversation and the service staff theme deviation value of the conversation.
17. The apparatus according to claim 16, wherein the second determining subunit is specifically configured to:
determining a semantic set subject to which a user question text of each response in the conversation belongs; determining deviation values between the semantic set topics to which the user question texts responded twice adjacently belong, and taking the determined deviation values as the deviation values of the user topics responded twice adjacently;
determining a semantic set subject to which a response text of a service person responding each time in the conversation belongs; and determining the deviation value between the topics of the semantic set to which the two adjacent responded server answer texts belong, and taking the determined deviation value as the topic deviation value of the two adjacent responded server answers.
18. The apparatus according to claim 16, wherein the second determining subunit is specifically configured to:
calculating the sum of the user theme deviation values of two adjacent responses in the conversation, and taking the calculation result as the user theme deviation value of the conversation;
and calculating the sum of the service personnel theme deviation values of two adjacent responses in the conversation, and taking the calculation result as the service personnel theme deviation value of the conversation.
19. The apparatus according to claim 16, wherein the second computing subunit is specifically configured to: and calculating a difference value between the user theme deviation value of the conversation and the service staff theme deviation value of the conversation, and taking a calculation result as the historical semantic deviation degree.
20. The apparatus according to any one of claims 12 to 19, wherein the second determining unit is specifically configured to: calculating a service offset value for the service person in the session according to the following formula:
QOP＝PerQ+∏HisQ；
wherein QEP represents a service offset value for a service person in the session; PerQ represents the current semantic deviation of the dialog; HisQ represents the historical semantic deviation of the dialog; ii denotes a configuration coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410470306.3A CN105404618B (en) | 2014-09-16 | 2014-09-16 | A kind of dialog text treating method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410470306.3A CN105404618B (en) | 2014-09-16 | 2014-09-16 | A kind of dialog text treating method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105404618A CN105404618A (en) | 2016-03-16 |
| CN105404618B true CN105404618B (en) | 2018-10-02 |
Family
ID=55470112
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410470306.3A Active CN105404618B (en) | 2014-09-16 | 2014-09-16 | A kind of dialog text treating method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105404618B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107862423B (en) * | 2017-06-28 | 2021-07-06 | 未鲲(上海)科技服务有限公司 | System evaluation method, intelligent evaluation system, and computer-readable storage medium |
| CN110796338A (en) * | 2019-09-24 | 2020-02-14 | 北京谦仁科技有限公司 | Online teaching monitoring method and device, server and storage medium |
| CN118521221A (en) * | 2024-07-17 | 2024-08-20 | 支付宝(杭州)信息技术有限公司 | Service dialogue quality inspection method and system and conversation quality estimation method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577556A (en) * | 2013-10-21 | 2014-02-12 | 北京奇虎科技有限公司 | Device and method for obtaining association degree of question and answer pair |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2616926A4 (en) * | 2010-09-24 | 2015-09-23 | Ibm | Providing question and answers with deferred type evaluation using text with limited structure |
-
2014
- 2014-09-16 CN CN201410470306.3A patent/CN105404618B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577556A (en) * | 2013-10-21 | 2014-02-12 | 北京奇虎科技有限公司 | Device and method for obtaining association degree of question and answer pair |
Non-Patent Citations (1)
| Title |
|---|
| 一种基于FAQ的自动应答系统的研究与设计;李敬伟等;《河南机电高等专科学校学报》;20120731;第20卷(第4期);27-30 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105404618A (en) | 2016-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107818798B (en) | Customer service quality evaluation method, device, equipment and storage medium | |
| CN108829682B (en) | Computer readable storage medium, intelligent question answering method and intelligent question answering device | |
| CN109002515A (en) | A kind of method and apparatus of intelligent response | |
| WO2020102093A1 (en) | System and method of assigning customer service tickets | |
| CN110580282B (en) | Method and device for interacting with customer service through simulation user | |
| CN110942326A (en) | User satisfaction prediction method and device and server | |
| CN113360622B (en) | User dialogue information processing method and device and computer equipment | |
| CN111049998A (en) | Voice customer service quality inspection method, customer service quality inspection equipment and storage medium | |
| CN112418779A (en) | Online self-service interviewing method based on natural language understanding | |
| CN105404618B (en) | A kind of dialog text treating method and apparatus | |
| CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
| CN114357126A (en) | Intelligent question-answering system | |
| CN116226355A (en) | Intelligent customer service method, system, electronic equipment and readable storage medium | |
| CN112288584A (en) | Insurance application processing method and device, computer readable medium and electronic equipment | |
| CN115050457A (en) | Method, device, equipment, medium and product for evaluating quality of on-line inquiry service | |
| CN110750620B (en) | Group decision capability evaluation method and device | |
| CN111400472A (en) | Legal consultation intelligent question-answering method and system | |
| CN110580899A (en) | Voice recognition method and device, storage medium and computing equipment | |
| CN111178982A (en) | Customer satisfaction analysis method, storage medium and computer device | |
| CN110765242A (en) | Method, device and system for providing customer service information | |
| CN111049999A (en) | Voice customer service quality inspection system and customer service quality inspection equipment | |
| CN114186048A (en) | Question-answer replying method and device based on artificial intelligence, computer equipment and medium | |
| KR20220123160A (en) | Artificial intelligence psychological status diagnostic system and methods | |
| CN113468297B (en) | Dialogue data processing method and device, electronic equipment and storage equipment | |
| CN115273854B (en) | Service quality determining method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200923 Address after: Grand Cayman Islands Patentee after: Innovative advanced technology Co.,Ltd. Address before: Grand Cayman Islands Patentee before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200923 Address after: Grand Cayman Islands Patentee after: Advanced innovation technology Co.,Ltd. Address before: Grand Cayman Islands Patentee before: Alibaba Group Holding Ltd. |
|
| TR01 | Transfer of patent right |