CN119669407A

CN119669407A - Data detection method, device, computer equipment and storage medium

Info

Publication number: CN119669407A
Application number: CN202411720023.XA
Authority: CN
Inventors: 张旭龙; 王健宗; 孔令炜
Original assignee: Ping An Technology Shenzhen Co Ltd
Current assignee: Ping An Technology Shenzhen Co Ltd
Priority date: 2024-11-27
Filing date: 2024-11-27
Publication date: 2025-03-21

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a data detection method, a device, computer equipment and a storage medium, wherein the method comprises the steps of acquiring dialogue text data which is collected in advance and corresponds to a target service; preprocessing dialogue text data to obtain target dialogue text, carrying out association rule mining on the target dialogue text based on an association rule mining algorithm to obtain initial association rules, screening appointed association rules related to emotion detection from the initial association rules, constructing a corresponding association rule set based on the appointed association rules, carrying out emotion rule mining on the association rule set based on a multi-target particle swarm optimization algorithm to obtain emotion rules, carrying out emotion violation detection on the dialogue text of target customer service based on the emotion rules, and obtaining a corresponding emotion violation detection result. In addition, emotion violation detection results may be stored in the blockchain. The method and the device are based on the use of emotion rules, and effectively improve the processing efficiency and the processing accuracy of emotion violation detection processing.

Description

Data detection method, device, computer equipment and storage medium

Technical Field

The present application relates to the field of artificial intelligence development technology and financial technology, and in particular, to a data detection method, apparatus, computer device, and storage medium.

Background

In the current digital transformation climax, cloud communication is taken as an enterprise-level communication service based on a cloud computing technology concept and a service mode, and is becoming a key tool for improving the operation efficiency and optimizing the customer experience of financial enterprises. The cloud communication service widely covers the traditional service fields of operators, such as short message service and voice service, and simultaneously goes deep into the core of internet service, and comprises the diversified functions of instant messaging I M, real-time audio and video communication, call center management, cloud customer service system, enterprise converged communication UC and the like. This trend not only greatly enriches the communication means of financial enterprises, but also significantly improves the flexibility and efficiency of communication.

It is particularly notable that with the rapid development of A I technology, the deep fusion with the customer service field is leading to the revolution of the customer service industry. The large number of repetitive and boring tasks that traditional customer service personnel undertake are now increasingly undertaken by intelligent customer service systems. The intelligent system can accurately convert voice information of a user into text through an advanced voice recognition (ASR) technology, then, the Natural Language Processing (NLP) technology is utilized to carry out intention recognition and matching on the words of the user, and finally, the matched reply content can be conveyed to the user in a voice form through a text-to-voice (TTS) technology by intelligent customer service, so that smooth and multi-turn voice interaction with the user is realized. The technical progress not only remarkably improves the efficiency of customer service work, but also brings more personalized and intelligent user experience.

However, while intelligent customer service is widely applied to the electric sales scenario, some problems to be solved urgently are also exposed. The quality of customer service attitude has a crucial influence on the success or failure of the electric marketing service. Traditional quality inspection methods often rely on manually analyzing call recordings of customer service to identify whether a violation of their emotion has occurred. However, this approach is not only inefficient, difficult to handle for large-scale, high-strength electrical pin business requirements, but also has a large uncertainty in accuracy.

Disclosure of Invention

The embodiment of the application aims to provide a data detection method, a data detection device, computer equipment and a storage medium, so as to solve the technical problems that the existing quality inspection mode of customer service emotion often depends on manual analysis of call recording of customer service, and has low efficiency and low accuracy.

In order to solve the above technical problems, the embodiment of the present application provides a data detection method, which adopts the following technical scheme:

acquiring dialogue text data which is collected in advance and corresponds to a target service;

Preprocessing the dialogue text data to obtain a corresponding target dialogue text;

performing association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain a corresponding initial association rule;

screening out a specified association rule related to emotion detection from the initial association rule;

constructing a corresponding association rule set based on the specified association rule;

carrying out emotion rule mining on the association rule set based on a multi-target particle swarm optimization algorithm to obtain a corresponding emotion rule;

And carrying out emotion violation detection on the dialogue text of the target customer service based on the emotion rule to obtain an emotion violation detection result corresponding to the target customer service.

Further, the step of performing association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain a corresponding initial association rule specifically includes:

Constructing a corresponding transaction database based on the target dialogue text;

Performing mining processing on the frequent item sets on the transaction database based on the association rule mining algorithm to obtain corresponding frequent item sets;

Generating a first association rule meeting a preset confidence threshold based on the frequent item set;

and taking the first association rule as the initial association rule.

Further, the step of screening the specified association rule related to emotion detection from the initial association rule specifically includes:

acquiring a preset index calculation strategy;

Calculating target indexes of all the initial association rules based on the index calculation strategy;

Screening all the initial association rules based on the target indexes to obtain second association rules conforming to emotion detection conditions;

And taking the second association rule as the appointed association rule.

Further, the step of constructing a corresponding association rule set based on the specified association rule specifically includes:

acquiring a preset target form;

adjusting the appointed association rule based on the target form to obtain a corresponding third association rule;

And integrating the third association rule to obtain the corresponding association rule set.

Further, the step of detecting emotion violation on the dialogue text of the target customer service based on the emotion rule to obtain an emotion violation detection result corresponding to the target customer service specifically includes:

Acquiring a dialogue text of the target customer service;

extracting attribute tags from the dialogue text to obtain corresponding attribute tags;

Extracting emotion labels from the dialogue text to obtain corresponding emotion labels;

matching the attribute tag with the emotion tag based on the emotion rule to obtain a corresponding matching result;

generating a score corresponding to the emotion rule based on the matching result;

judging whether the score is larger than a preset score threshold value or not;

And if the score is larger than the score threshold, generating a first emotion violation detection result of emotion violations of the target customer service in the dialogue text, otherwise, generating a second emotion violation detection result of emotion violations of the target customer service in the dialogue text.

Further, the step of preprocessing the dialog text data to obtain a corresponding target dialog text specifically includes:

Word segmentation processing is carried out on the dialogue text data to obtain a corresponding first dialogue text;

Removing irrelevant information from the first dialogue text to obtain a corresponding second dialogue text;

carrying out standardization processing on the second dialogue text to obtain a corresponding third dialogue text;

and taking the third dialogue text as the target dialogue text.

Further, after the step of detecting emotion violation on the dialogue text of the target customer service based on the emotion rule to obtain an emotion violation detection result corresponding to the target customer service, the method further includes:

judging whether the emotion violation detection result is that whether the target customer service has emotion violations in the dialogue text or not;

if yes, obtaining violation information corresponding to the target customer service;

generating a corresponding violation report based on the violation information;

And sending the violation report to related personnel.

In order to solve the above technical problems, the embodiment of the present application further provides a data detection device, which adopts the following technical scheme:

the first acquisition module is used for acquiring dialogue text data which is collected in advance and corresponds to the target service;

The preprocessing module is used for preprocessing the dialogue text data to obtain a corresponding target dialogue text;

the first mining module is used for carrying out association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain a corresponding initial association rule;

The screening module is used for screening out appointed association rules related to emotion detection from the initial association rules;

the construction module is used for constructing a corresponding association rule set based on the appointed association rule;

The second mining module is used for mining the emotion rules of the association rule set based on a multi-target particle swarm optimization algorithm to obtain corresponding emotion rules;

And the detection module is used for carrying out emotion violation detection on the dialogue text of the target customer service based on the emotion rule to obtain an emotion violation detection result corresponding to the target customer service.

In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical schemes:

In order to solve the above technical problems, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical schemes:

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

The method comprises the steps of firstly obtaining dialogue text data which are collected in advance and correspond to target business, preprocessing the dialogue text data to obtain corresponding target dialogue texts, then carrying out association rule mining on the target dialogue texts based on a preset association rule mining algorithm to obtain corresponding initial association rules, screening appointed association rules related to emotion detection from the initial association rules, constructing a corresponding association rule set based on the appointed association rules, carrying out emotion rule mining on the association rule set based on a multi-target particle swarm optimization algorithm to obtain corresponding emotion rules, and finally carrying out emotion violation detection on dialogue texts of target customer service based on the emotion rules to obtain emotion violation detection results corresponding to the target customer service. According to the method, the target dialogue text is obtained by preprocessing dialogue text data which are collected in advance and correspond to the target business, then the target dialogue text is subjected to association rule mining based on the use of an association rule mining algorithm to obtain an initial association rule, then the appointed association rule related to emotion detection is screened out from the initial association rule, and an association rule set is constructed, and further emotion rule mining is carried out on the association rule set based on the use of a multi-target particle swarm optimization algorithm to obtain emotion rules, so that emotion rule detection processing of the dialogue text of the target customer service can be effectively and accurately carried out based on the use of the emotion rules, the processing efficiency and the processing accuracy of the emotion rule detection processing are effectively improved, and the accuracy of the obtained emotion rule detection result is ensured.

Drawings

In order to more clearly illustrate the solution of the present application, a brief description will be given below of the drawings required for the description of the embodiments of the present application, it being apparent that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without the exercise of inventive effort for a person of ordinary skill in the art.

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of one embodiment of a data detection method according to the present application;

FIG. 3 is a schematic diagram of a data detection device according to one embodiment of the present application;

FIG. 4 is a schematic structural diagram of one embodiment of a computer device in accordance with the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs, the terms used in the description herein are used for the purpose of describing particular embodiments only and are not intended to limit the application, and the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the above description of the drawings are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

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.

In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, the system architecture 100 may include a terminal device 101, a network 102, and a server 103, where the terminal device 101 may be a notebook 1011, a tablet 1012, or a cell phone 1013. Network 102 is the medium used to provide communication links between terminal device 101 and server 103. Network 102 may include various connection types such as wired, wireless communication links, or fiber optic cables, among others.

A user may interact with the server 103 via the network 102 using the terminal device 101 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, and the like, may be installed on the terminal device 101.

The terminal device 101 may be various electronic devices having a display screen and supporting web browsing, and the terminal device 101 may be an electronic book reader, an MP3 player (Moving Picture Experts Group Audio Layer III, moving picture experts compression standard audio layer III), an MP4 (Moving Picture Experts Group Audio Layer IV, moving picture experts compression standard audio layer IV) player, a laptop portable computer, a desktop computer, and the like, in addition to the notebook 1011, the tablet 1012, or the mobile phone 1013.

The server 103 may be a server providing various services, such as a background server providing support for pages displayed on the terminal device 101.

It should be noted that, the data detection method provided by the embodiment of the present application is generally executed by a server/terminal device, and accordingly, the data detection device is generally disposed in the server/terminal device.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow chart of one embodiment of a data detection method according to the present application is shown. The order of the steps in the flowchart may be changed and some steps may be omitted according to various needs. The data detection method provided by the embodiment of the application can be applied to any scene needing data detection, and can be applied to products of the scenes, such as data detection of customer service emotion in the field of financial insurance. The data detection method comprises the following steps:

step S201, dialogue text data corresponding to the target service collected in advance is acquired.

In this embodiment, the electronic device (for example, the server/terminal device shown in fig. 1) on which the data detection method operates may acquire the dialogue text data corresponding to the target service through a wired connection manner or a wireless connection manner. It should be noted that the wireless connection may include, but is not limited to, 3G/4G/5G connection, wiFi connection, bluetooth connection, wiMAX connection, zigbee connection, UWB (ultra wideband) connection, and other now known or later developed wireless connection. The execution subject of the application can be specifically an intelligent customer service system or simply a system. The target service may be specifically an electric pin service. Specifically, a log file containing customer service and user dialogue content can be exported through a background interface of the intelligent customer service system. These files typically include a timestamp of the conversation, a customer service ID, a user ID, and specific conversation text. And the exported log files are sorted, invalid data (such as empty dialogue, system prompt and the like) are removed, and the accuracy and the integrity of the data are ensured, so that corresponding dialogue text data are obtained.

Step S202, preprocessing the dialogue text data to obtain a corresponding target dialogue text.

In this embodiment, the above preprocessing is performed on the dialog text data to obtain a specific implementation process of the corresponding target dialog text, which will be described in further detail in the following specific embodiments, which will not be described herein.

Step S203, performing association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain a corresponding initial association rule.

In this embodiment, the foregoing association rule mining algorithm performs association rule mining on the target dialog text to obtain a specific implementation process of a corresponding initial association rule, which will be described in further detail in the following specific embodiments, which will not be described in any more detail herein.

Step S204, a designated association rule relevant to emotion detection is screened out from the initial association rules.

In this embodiment, the specific implementation process of selecting the specific association rule related to emotion detection from the initial association rules is described in further detail in the following specific embodiments, which will not be described herein.

Step S205, constructing a corresponding association rule set based on the specified association rule.

In this embodiment, the specific implementation process of constructing the corresponding association rule set based on the specified association rule is described in further detail in the following specific embodiment, which is not described herein.

And S206, carrying out emotion rule mining on the association rule set based on a multi-target particle swarm optimization algorithm to obtain a corresponding emotion rule.

In this embodiment, the process of performing emotion rule mining on the association rule set based on the multi-objective particle swarm optimization algorithm to obtain the corresponding emotion rule includes the steps of initializing a particle swarm, defining a fitness function, updating particle speed and position, performing iterative search, forming the emotion rule, and the like. Specifically, initializing a population of particles includes sizing the population of particles, each particle representing a set of possible emotion rule parameters. These parameters may include look-ahead conditions in the association rules (e.g., user-mentioned emotion words), outcome conditions (e.g., customer-presented emotion states), and weights of rules, etc. Defining the fitness function comprises the step of evaluating the merits of the emotion rules represented by the particles by the fitness function. The method can calculate the indexes of accuracy, recall rate and the like of the rule represented by each particle in detecting emotion violations based on historical dialogue data, and integrate the indexes to form a fitness value. Updating the particle velocity and position includes updating the velocity and position of the particle based on the historical best position (individual best and global best) and the current particle velocity and position. This process simulates the process of a particle finding the optimal solution in the search space. Iterative searching involves iteratively finding optimal parameters in the search space until a stopping condition is met (e.g., a maximum number of iterations is reached or a solution is found that meets the requirements). In the iterative process, the global best solution and the individual best solution are continuously recorded and updated. Forming emotion rules includes forming a specific plurality of emotion rules based on the resulting particles (i.e., optimal solutions). The emotion rules may include emotion states that customer service may exhibit under certain circumstances and associations between the emotion states and user feedback.

And step S207, emotion violation detection is carried out on the dialogue text of the target customer service based on the emotion rule, and an emotion violation detection result corresponding to the target customer service is obtained.

In this embodiment, the emotion violation detection is performed on the dialogue text of the target customer service based on the emotion rule, so as to obtain a specific implementation process of the emotion violation detection result corresponding to the target customer service.

In some alternative implementations, step S203 includes the steps of:

And constructing a corresponding transaction database based on the target dialogue text.

In this embodiment, the dialogue text is divided by sentences or dialogue turns, each of which is regarded as a transaction. And further, words in the transactions are regarded as items, and a corresponding transaction database is constructed, wherein each transaction comprises a group of items (namely words).

And carrying out mining processing on the frequent item sets on the transaction database based on the association rule mining algorithm to obtain corresponding frequent item sets.

In this embodiment, the association rule mining algorithm may specifically use Apriori algorithm or FP-Growth algorithm. After determining the selection of the association rule mining algorithm, frequent item sets in the transaction database may be mined by using the association rule mining algorithm.

And generating a first association rule meeting a preset confidence threshold value based on the frequent item set.

In this embodiment, the confidence threshold is not specifically limited, and may be set according to actual service requirements. The first association rule satisfying the confidence threshold represents a meaningful association rule.

And taking the first association rule as the initial association rule.

The method comprises the steps of constructing a corresponding transaction database based on the target dialogue text, then carrying out mining processing on a frequent item set on the transaction database based on the association rule mining algorithm to obtain a corresponding frequent item set, then generating a first association rule meeting a preset confidence threshold based on the frequent item set, and subsequently taking the first association rule as the initial association rule. According to the method, the corresponding transaction database is constructed based on the target dialogue text, then the mining processing of the frequent item set is carried out on the transaction database based on the use of the association rule mining algorithm to obtain the corresponding frequent item set, and further the initial association rule meeting the confidence coefficient threshold can be efficiently and accurately generated based on the frequent item set, so that the generation efficiency of the initial association rule is improved, and the accuracy of the obtained initial association rule is ensured.

In some alternative implementations of the present embodiment, step S204 includes the steps of:

And acquiring a preset index calculation strategy.

In this embodiment, the index calculation policy includes a policy for calculating an index corresponding to a confidence level, a coverage level, and an interest level of the association rule. The association rule is stated in the form of "if-then", the condition associated in the "if" section is called a preceding condition, and the condition associated in the "then" section is called a result condition, respectively called a and C. The association rule is expressed as R.A.fwdarw.C. The support count of the association rule is denoted by SUP (a→c), which is the number of transactions compatible with a and C at the same time, i.e., the number of transactions containing one a→c, SUP (a→c) =sup (a→c). Similarly, SUP (a) and SUP (C) are the number of transactions compatible only with a and C, respectively.

Specifically, the confidence of an association rule is used to measure specificity or consistency, representing the probability of creating a rule depending on the antecedent portion, defined as follows:

The coverage of an association rule is the extent to which the measurement result portion is covered by the association rule, which shows the likelihood of creating a rule based on the result portion. The coverage is defined as follows:

the interestingness of the association rule is a measure of how surprising the association rule is to the user. The definition of interest is as follows: wherein N is a natural number, and the specific value can be adjusted according to historical data, expert experience or actual requirements.

And calculating target indexes of all the initial association rules based on the index calculation strategy.

In this embodiment, the target index of all the initial association rules may be calculated according to the policy content of the index calculation policy, that is, the confidence level, coverage level and interest level of all the initial association rules are included.

And screening all the initial association rules based on the target index to obtain a second association rule conforming to emotion detection conditions.

In this embodiment, a confidence threshold, a coverage threshold, and an interest threshold corresponding to the confidence, coverage, and interest, respectively, are preset. And traversing all the mined initial association rules, and screening each initial association rule according to the calculated target indexes of each initial association rule so as to keep the rules meeting all index thresholds (confidence threshold, coverage threshold and interestingness threshold) at the same time, so as to obtain a second association rule meeting emotion detection conditions.

And taking the second association rule as the appointed association rule.

The method comprises the steps of obtaining a preset index calculation strategy, calculating target indexes of all initial association rules based on the index calculation strategy, screening all the initial association rules based on the target indexes to obtain second association rules meeting emotion detection conditions, and taking the second association rules as the appointed association rules. According to the method, the target indexes of all initial association rules are calculated based on the use of the index calculation strategy, and then all initial association rules are screened based on the use of the target indexes, so that the appointed association rules related to emotion detection can be screened out quickly and accurately, the screening efficiency of the appointed association rules is improved, and the accuracy of the obtained appointed association rules is ensured.

In some alternative implementations, step S205 includes the steps of:

And acquiring a preset target form.

In the present embodiment, the above-described target forms may specifically include a form representation of "if-then", in which the "if" portion represents the preceding condition (a) and the "then" portion represents the result condition (C), forming the rule R: a→c.

And adjusting the specified association rule based on the target form to obtain a corresponding third association rule.

In this embodiment, the specified association rule obtained after the filtering may be represented, so as to obtain the adjusted third association rule.

In the present embodiment, the above-described association rule set is a set made up of all the third association rules.

The method comprises the steps of obtaining a preset target form, then adjusting the appointed association rule based on the target form to obtain a corresponding third association rule, and then integrating the third association rule to obtain the corresponding association rule set. According to the method, the corresponding third association rule is obtained by adjusting the designated association rule based on the obtained target form, and then the third association rule is integrated, so that the corresponding association rule set can be quickly and accurately constructed, and the construction efficiency and the construction accuracy of the association rule set are improved.

In some alternative implementations, step S207 includes the steps of:

And acquiring the dialogue text of the target customer service.

In this embodiment, the dialogue text of the target customer service is a dialogue text that needs emotion violation processing.

And extracting the attribute tags of the dialogue text to obtain corresponding attribute tags.

In this embodiment, the text classification algorithm may be used to classify the dialogue text of the target customer service, so as to extract attribute tags (such as customer service attitudes, user feedback, etc.) corresponding to the dialogue text.

And extracting the emotion label from the dialogue text to obtain a corresponding emotion label.

In this embodiment, emotion labels corresponding to the dialogue text can be extracted by performing emotion analysis on the dialogue text of the target customer service by using a pre-trained emotion analysis model or a custom emotion dictionary.

And carrying out matching processing on the attribute tag and the emotion tag based on the emotion rule to obtain a corresponding matching result.

In this embodiment, the number of emotion rules may include a plurality of emotion rules. The matching process includes traversing all emotion rules. For each emotion rule, it is checked whether its antecedent matches the extracted emotion tag and attribute tag. If the matching is successful, the emotion rules and relevant information thereof are recorded, and the matching result of the emotion rules is that the matching is successful.

And generating a score corresponding to the emotion rule based on the matching result.

In this embodiment, each rule successfully matched may be screened out according to the matching result, and the rule score of the rule may be calculated according to the weight of the rule, so as to perform weighted summation on all rule scores or perform calculation by using more complex calculation methods (such as weighted product, weighted average, etc.), thereby obtaining the aggregate score of all rules successfully matched, that is, the score. The weights of the rules can be set according to actual service requirements.

And judging whether the score is larger than a preset score threshold value or not.

In this embodiment, the threshold of the emotion rule score may be set according to the actual requirement. This threshold may be adjusted based on historical data, expert experience, or actual demand.

In this embodiment, the emotion rule total score of the dialog text, that is, the score is compared with the score threshold. If the score exceeds the score threshold, the condition that the target customer service possibly has emotion violations in the dialogue is considered, and then a first emotion violation detection result that the target customer service has emotion violations in the dialogue text is generated. And if the score does not exceed the score threshold, the condition that the target customer service does not have emotion violations in the dialogue is considered, and a second emotion violation detection result that the target customer service does not have emotion violations in the dialogue text is generated.

The method comprises the steps of obtaining a dialogue text of a target customer service, extracting attribute tags of the dialogue text to obtain corresponding attribute tags, extracting emotion tags of the dialogue text to obtain corresponding emotion tags, carrying out matching processing on the attribute tags and the emotion tags based on the emotion rules to obtain corresponding matching results, generating scores corresponding to the emotion rules based on the matching results, finally judging whether the scores are larger than a preset score threshold, generating a first emotion violation detection result of emotion violations of the target customer service in the dialogue text if the scores are larger than the score threshold, and generating a second emotion violation detection result of emotion violations of the target customer service in the dialogue text if the scores are not larger than the score threshold. According to the method, the attribute label and the emotion label are extracted from the dialogue text of the target customer service to obtain the corresponding attribute label and emotion label, the attribute label and the emotion label are matched based on the use of emotion rules to obtain a matching result, the score corresponding to the emotion rules is generated based on the matching result, and the score is compared with the preset score threshold value, so that emotion violation detection processing of the dialogue text of the target customer service can be rapidly and accurately completed according to the obtained value comparison result, and the accuracy and reliability of emotion violation detection processing are effectively ensured.

In some alternative implementations of the present embodiment, step S202 includes the steps of:

and performing word segmentation processing on the dialogue text data to obtain a corresponding first dialogue text.

In this embodiment, the word segmentation processing may be performed on the dialog text data by using a natural language processing library, so as to obtain the corresponding first dialog text.

And removing irrelevant information from the first dialogue text to obtain a corresponding second dialogue text.

In this embodiment, the above-mentioned removal-independent information processing includes removal of stop words and removal of punctuation marks. Wherein, the removing the stop words includes: common but not practical words (e.g. "have", etc.) in the dialog text are removed from the stop vocabulary. Removing punctuation includes removing punctuation in the dialog text using regular expressions or string processing methods.

And carrying out standardization processing on the second dialogue text to obtain a corresponding third dialogue text.

In the present embodiment, the above normalization processing includes processing of converting uppercase letters into lowercase letters, unifying punctuation marks, and the like, to be converted into a normalized word sequence.

And taking the third dialogue text as the target dialogue text.

The method comprises the steps of performing word segmentation on dialogue text data to obtain corresponding first dialogue texts, removing irrelevant information from the first dialogue texts to obtain corresponding second dialogue texts, performing standardization processing on the second dialogue texts to obtain corresponding third dialogue texts, and taking the third dialogue texts as target dialogue texts. According to the method, the device and the system, word segmentation processing, irrelevant information removal processing and standardization processing are carried out on the dialogue text data, so that preprocessing of the dialogue text data can be efficiently and accurately completed, and the accuracy of the obtained target dialogue text is ensured.

In some optional implementations of this embodiment, after step S207, the electronic device may further perform the following steps:

Judging whether the emotion violation detection result is that whether the target customer service has emotion violations in the dialogue text or not.

In this embodiment, the content of the emotion violation detection result may include that the target customer service has emotion violations in the dialogue text, or that the target customer service does not have emotion violations in the dialogue text.

If yes, obtaining the violation information corresponding to the target customer service.

In this embodiment, the above-mentioned violation information includes information such as ID of the target customer service, violation time, violation content, and the like.

And generating a corresponding violation report based on the violation information.

In this embodiment, the preset violation report template may be obtained, and then the violation information may be filled into the corresponding position in the violation report template, so as to generate a corresponding violation report. The content of the violation report template can be constructed according to actual service requirements. For example, the violation reporting template may include an ID area, a time area, and a content area.

And sending the violation report to related personnel.

In this embodiment, the related person may be a customer service manager, or may also be a person responsible for investigation.

The method comprises the steps of judging whether the emotion violation detection result is that emotion violations exist in the dialogue text for the target customer service or not, if so, obtaining violation information corresponding to the target customer service, then generating a corresponding violation report based on the violation information, and then sending the violation report to related personnel. According to the method and the device, when the emotion violation detection result is detected to be that the emotion violation exists in the dialogue text by the target customer service, the violation information corresponding to the target customer service can be automatically acquired, and the corresponding violation report is generated based on the violation information, so that the generation efficiency and the generation intelligence of the violation report are improved. And the follow-up report is sent to related personnel, so that the related personnel can carry out corresponding internal investigation and processing on target customer service according to the report, and the use experience of the related personnel is improved.

In some alternative implementations, the obtained user information solicits user consent and meets the specifications of the relevant laws and relevant policies.

In addition, the non-native company software tools or components present in the embodiments of the present application are presented by way of example only and are not representative of actual use.

It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.

It should be emphasized that, to further ensure the privacy and security of the emotion violation detection results, the emotion violation detection results may also be stored in a node of a blockchain.

The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The blockchain (Blockchain), essentially a de-centralized database, is a string of data blocks that are generated in association using cryptographic methods, each of which contains information from a batch of network transactions for verifying the validity (anti-counterfeit) of its information and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Wherein artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) is the theory, method, technique, and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend, and expand human intelligence, sense the environment, acquire knowledge, and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by computer readable instructions stored in a computer readable storage medium that, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

With further reference to fig. 3, as an implementation of the method shown in fig. 2, the present application provides an embodiment of a data detection apparatus, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus is particularly applicable to various electronic devices.

As shown in fig. 3, the data detection apparatus 300 in this embodiment includes a first acquisition module 301, a preprocessing module 302, a first mining module 303, a screening module 304, a construction module 305, a second mining module 306, and a detection module 307. Wherein:

A first obtaining module 301, configured to obtain dialogue text data collected in advance and corresponding to a target service;

The preprocessing module 302 is configured to preprocess the dialog text data to obtain a corresponding target dialog text;

The first mining module 303 is configured to perform association rule mining on the target dialog text based on a preset association rule mining algorithm, so as to obtain a corresponding initial association rule;

a screening module 304, configured to screen out a specified association rule related to emotion detection from the initial association rules;

A construction module 305, configured to construct a corresponding association rule set based on the specified association rule;

the second mining module 306 is configured to perform emotion rule mining on the association rule set based on a multi-objective particle swarm optimization algorithm to obtain a corresponding emotion rule;

and the detection module 307 is configured to perform emotion violation detection on the dialogue text of the target customer service based on the emotion rule, so as to obtain an emotion violation detection result corresponding to the target customer service.

In this embodiment, the operations performed by the modules or units respectively correspond to the steps of the data detection method in the foregoing embodiment one by one, and are not described herein again.

In some alternative implementations of the present embodiment, the first mining module 303 includes:

The construction sub-module is used for constructing a corresponding transaction database based on the target dialogue text;

the mining sub-module is used for mining the frequent item sets on the transaction database based on the association rule mining algorithm to obtain the corresponding frequent item sets;

a first generation sub-module, configured to generate a first association rule that satisfies a preset confidence threshold based on the frequent item set;

and the first determining submodule is used for taking the first association rule as the initial association rule.

In some alternative implementations of the present embodiment, the screening module 304 includes:

The first acquisition submodule is used for acquiring a preset index calculation strategy;

a calculation sub-module, configured to calculate target indexes of all the initial association rules based on the index calculation policy;

The first processing submodule is used for screening all the initial association rules based on the target index to obtain a second association rule which accords with emotion detection conditions;

And the second determining submodule is used for taking the second association rule as the appointed association rule.

In some alternative implementations of the present embodiment, the building block 305 includes:

The second acquisition sub-module is used for acquiring a preset target form;

The adjustment sub-module is used for adjusting the specified association rule based on the target form to obtain a corresponding third association rule;

And the integration sub-module is used for integrating the third association rule to obtain the corresponding association rule set.

In some alternative implementations of the present embodiment, the detection module 307 includes:

A third obtaining sub-module, configured to obtain a dialogue text of the target customer service;

The first extraction sub-module is used for extracting attribute tags of the dialogue text to obtain corresponding attribute tags;

The second extraction submodule is used for extracting emotion labels from the dialogue text to obtain corresponding emotion labels;

the matching sub-module is used for carrying out matching processing on the attribute tag and the emotion tag based on the emotion rule to obtain a corresponding matching result;

the second generation sub-module is used for generating a score corresponding to the emotion rule based on the matching result;

The judging submodule is used for judging whether the score is larger than a preset score threshold value or not;

And the third generation sub-module is used for generating a first emotion violation detection result of emotion violations of the target customer service in the dialogue text if the score is larger than the score threshold, and generating a second emotion violation detection result of emotion violations of the target customer service in the dialogue text if the score is not larger than the score threshold.

In some alternative implementations of the present embodiment, the preprocessing module 302 includes:

the second processing sub-module is used for carrying out word segmentation processing on the dialogue text data to obtain a corresponding first dialogue text;

the third processing sub-module is used for removing irrelevant information processing on the first dialogue text to obtain a corresponding second dialogue text;

A fourth processing sub-module, configured to perform standardization processing on the second dialog text, so as to obtain a corresponding third dialog text;

and the third determining submodule is used for taking the third dialogue text as the target dialogue text.

In some optional implementations of this embodiment, the data detection apparatus further includes:

the judging module is used for judging whether the emotion violation detection result is that whether emotion violations exist in the dialogue text for the target customer service or not;

the second acquisition module is used for acquiring the violation information corresponding to the target customer service if yes;

the generation module is used for generating a corresponding violation report based on the violation information;

and the sending module is used for sending the violation report to related personnel.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 4, fig. 4 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It should be noted that only computer device 4 having components 41-43 is shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and its hardware includes, but is not limited to, a microprocessor, an Application SPECIFIC INTEGRATED Circuit (ASIC), a Programmable gate array (Field-Programmable GATE ARRAY, FPGA), a digital Processor (DIGITAL SIGNAL Processor, DSP), an embedded device, and the like.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 41 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the computer device 4. Of course, the memory 41 may also comprise both an internal memory unit of the computer device 4 and an external memory device. In this embodiment, the memory 41 is typically used to store an operating system and various application software installed on the computer device 4, such as computer readable instructions of a data detection method. Further, the memory 41 may be used to temporarily store various types of data that have been output or are to be output.

The processor 42 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions stored in the memory 41 or process data, such as computer readable instructions for executing the data detection method.

The network interface 43 may comprise a wireless network interface or a wired network interface, which network interface 43 is typically used for establishing a communication connection between the computer device 4 and other electronic devices.

According to the embodiment of the application, the target dialogue text is obtained by preprocessing dialogue text data which is collected in advance and corresponds to the target business, then the target dialogue text is subjected to association rule mining based on the use of an association rule mining algorithm to obtain an initial association rule, then the appointed association rule related to emotion detection is screened out from the initial association rule, and an association rule set is constructed, and further the emotion rule mining is carried out on the association rule set based on the use of a multi-target particle swarm optimization algorithm to obtain the emotion rule, so that the emotion rule based on the use of the emotion rule can be used subsequently, the emotion violation detection processing of the dialogue text of the target customer service can be effectively and accurately carried out, the processing efficiency and the processing accuracy of the emotion violation detection processing are effectively improved, and the accuracy of the obtained emotion violation detection result is ensured.

The present application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the data detection method as described above.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

It is apparent that the above-described embodiments are only some embodiments of the present application, but not all embodiments, and the preferred embodiments of the present application are shown in the drawings, which do not limit the scope of the patent claims. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a thorough and complete understanding of the present disclosure. Although the application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing description, or equivalents may be substituted for elements thereof. All equivalent structures made by the content of the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the scope of the application.

Claims

1. A data detection method, characterized in that it comprises the following steps:

Acquire pre-collected conversation text data corresponding to the target business;

Performing association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain corresponding initial association rules;

Filtering out designated association rules related to emotion detection from the initial association rules;

Based on a multi-objective particle swarm optimization algorithm, sentiment rules are mined on the association rule set to obtain corresponding sentiment rules;

Based on the emotion rule, emotion violation detection is performed on the conversation text of the target customer service to obtain an emotion violation detection result corresponding to the target customer service.

2. The data detection method according to claim 1 is characterized in that the step of performing association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain the corresponding initial association rules specifically comprises:

Building a corresponding transaction database based on the target dialogue text;

Performing frequent item set mining processing on the transaction database based on the association rule mining algorithm to obtain corresponding frequent item sets;

Generating a first association rule that meets a preset confidence threshold based on the frequent item set;

The first association rule is used as the initial association rule.

3. The data detection method according to claim 1, characterized in that the step of selecting the designated association rules related to emotion detection from the initial association rules specifically comprises:

Get the preset indicator calculation strategy;

Calculate the target indicators of all the initial association rules based on the indicator calculation strategy;

Screening all the initial association rules based on the target index to obtain a second association rule that meets the emotion detection condition;

The second association rule is used as the designated association rule.

4. The data detection method according to claim 1, characterized in that the step of constructing a corresponding association rule set based on the specified association rule specifically comprises:

Get the preset target form;

Adjusting the specified association rule based on the target form to obtain a corresponding third association rule;

The third association rules are integrated to obtain the corresponding association rule set.

5. The data detection method according to claim 1, characterized in that the step of performing emotion violation detection on the conversation text of the target customer service based on the emotion rule to obtain the emotion violation detection result corresponding to the target customer service specifically comprises:

Obtaining the conversation text of the target customer service;

Extracting attribute tags from the conversation text to obtain corresponding attribute tags;

Extracting emotion tags from the conversation text to obtain corresponding emotion tags;

Based on the emotion rule, the attribute label and the emotion label are matched to obtain a corresponding matching result;

Generating a score corresponding to the sentiment rule based on the matching result;

Determining whether the score is greater than a preset score threshold;

If the score is greater than the score threshold, a first emotion violation detection result is generated indicating that the target customer service has an emotion violation in the conversation text; otherwise, a second emotion violation detection result is generated indicating that the target customer service has no emotion violation in the conversation text.

6. The data detection method according to claim 1, characterized in that the step of preprocessing the dialogue text data to obtain the corresponding target dialogue text specifically comprises:

Performing word segmentation processing on the dialogue text data to obtain a corresponding first dialogue text;

performing standardization processing on the second dialogue text to obtain a corresponding third dialogue text;

The third dialogue text is used as the target dialogue text.

7. The data detection method according to claim 1, characterized in that after the step of performing emotion violation detection on the conversation text of the target customer service based on the emotion rule to obtain the emotion violation detection result corresponding to the target customer service, it also includes:

Determine whether the content of the emotion violation detection result indicates that the target customer service has an emotion violation in the conversation text;

If so, obtain the violation information corresponding to the target customer service;

Generate a corresponding violation report based on the violation information;

Send said violation report to the appropriate personnel.

8. A data detection device, comprising:

A first acquisition module, used to acquire pre-collected conversation text data corresponding to the target business;

A preprocessing module, used for preprocessing the dialogue text data to obtain a corresponding target dialogue text;

A first mining module is used to perform association rule mining on the target dialogue text based on a preset association rule mining algorithm to obtain corresponding initial association rules;

A screening module, used to screen out designated association rules related to emotion detection from the initial association rules;

A construction module, used to construct a corresponding association rule set based on the specified association rule;

The second mining module is used to mine sentiment rules on the association rule set based on a multi-objective particle swarm optimization algorithm to obtain corresponding sentiment rules;

The detection module is used to perform emotion violation detection on the conversation text of the target customer service based on the emotion rule to obtain an emotion violation detection result corresponding to the target customer service.

9. A computer device, comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the data detection method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that computer-readable instructions are stored on the computer-readable storage medium, and when the computer-readable instructions are executed by a processor, the steps of the data detection method according to any one of claims 1 to 7 are implemented.