CN113112282A - Method, device, equipment and medium for processing consult problem based on client portrait - Google Patents
Method, device, equipment and medium for processing consult problem based on client portrait Download PDFInfo
- Publication number
- CN113112282A CN113112282A CN202110423425.3A CN202110423425A CN113112282A CN 113112282 A CN113112282 A CN 113112282A CN 202110423425 A CN202110423425 A CN 202110423425A CN 113112282 A CN113112282 A CN 113112282A
- Authority
- CN
- China
- Prior art keywords
- client
- consult
- information
- portrait
- processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012545 processing Methods 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000004458 analytical method Methods 0.000 claims abstract description 35
- 230000009193 crawling Effects 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 32
- 239000013598 vector Substances 0.000 claims description 19
- 230000006399 behavior Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000008901 benefit Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000013016 damping Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Evolutionary Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the field of artificial intelligence, in particular to a method, a device, equipment and a medium for processing consult questions based on client figures in the field of classification. The invention comprises the following steps: acquiring client consult information, and performing semantic analysis on the client consult information to obtain consult keywords; crawling a client first-level label representing client behavior information and basic information through network data; and inputting the consult keywords and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label. The consultative processing advice is generated by the client portrait and the client secondary label to enable consultative processing staff to more comprehensively and quickly know the information such as the personality, hobbies, academic calendar, potential requirements and the like of the client, so that the requirement of the client work order is more accurately grasped, the speed of processing the work order is increased, the time for processing the work order is shortened, the time for visiting the client and analyzing the problems by the telephone is reduced, the satisfaction degree of the client is improved, and the labor is saved.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a method, a device, equipment and a medium for processing consult questions based on client figures in the field of classification.
Background
With the rapid development of network technology, the way of handling business on the internet is more and more popular. When conducting online business transaction, a client usually consults an electronic consult for problems such as service attitude, business cost, etc. of customer service, and further consults a superior supervision department if the result of response or processing is unsatisfactory. The existing processing method is to input the customer's appeal into the consulting work order system in the form of a work order through a seat. Customer work orders are typically processed through these two main steps:
1. and understanding the work order to be processed according to the content, the emergency degree, the service category and other information of the work order.
2. The telephone client further learns the appeal of the confirmed client and processes the work order.
Due to the difference in seat understanding, the quality of the entered work orders is uneven, the work orders often cannot accurately reflect the requirements of the clients, and the consulting staff usually does not know the specific conditions of the clients and is not instructive to return to the clients, so that the results of the work order processing are poor in experience for the clients, and a large amount of manpower and time analysis problems and return to the clients are wasted.
Disclosure of Invention
Based on the above technical problems, a method, a device, equipment and a medium for processing consultations and questions based on client portrait are provided, when a consultations work order of a client needs to be processed, the client portrait of the client and the consultations and processing advice generated based on a client secondary label are visually displayed to consultations and processing personnel, so that the consultations and processing personnel can more comprehensively and quickly know the information of the target client such as the character, hobbies, academic history, potential requirements and the like, the requirements of the client work order can be more accurately grasped, the speed of processing the work order is accelerated, the work order processing timeliness is shortened, the time for telephone return visit to the client and problem analysis is reduced, the client satisfaction is improved, and the labor is saved.
A method for processing a consult question based on a client portrait comprises the following steps:
acquiring client consult information, and performing semantic analysis on the client consult information to obtain consult key words;
crawling a client first-level label representing client behavior information and basic information through network data;
and inputting the consulting keyword and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label.
An apparatus for processing a consult question based on a client portrait, comprising:
the key word module is used for acquiring client consult information and performing semantic analysis on the client consult information to obtain consult keywords;
the first-level label module is used for crawling a first-level label of the client representing the behavior information and the basic information of the client through network data;
and the secondary label module is used for inputting the consult keyword and the client primary label into a client portrait model for analysis to generate a client portrait and obtain a client secondary label.
A computer apparatus comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, the processor when executing the computer readable instructions implementing the method for processing a consult problem based on a client representation.
One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the method for processing the problem based on the client representation as described above.
The method, the device, the equipment and the medium for processing the consult question based on the client portrait acquire client consult information, perform semantic analysis on the client consult information to acquire consult keywords and determine the keywords of the client consult; further, a client level label representing client behavior information and basic information is crawled through network data, and more existing information data of a client are obtained; and inputting the consulting keyword and the client first-level label into a client portrait model for analysis, generating a client portrait, and obtaining a client second-level label. The consultative consultation treatment suggestion is generated by the client portrait and based on the client secondary label, so that consultative consultation treatment staff can make treatment judgment on consultative consultation treatment problems more comprehensively and quickly, the requirement of a target client work order is more accurately grasped, the speed of work order treatment is accelerated, the work order treatment time is shortened, the time for visiting clients and analyzing problems by telephone is reduced, the client satisfaction degree is improved, and labor is saved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced 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 that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of a method for processing a consult question based on a client portrait in an embodiment of the present invention;
FIG. 2 is a schematic view of a process for processing a consult question based on a client portrait in an embodiment of the present invention;
FIG. 3 is a schematic view of a process for processing a consult question based on a client portrait in an embodiment of the present invention;
FIG. 4 is a schematic view of a process for processing a consult question based on a client portrait in an embodiment of the present invention;
FIG. 5 is a schematic view of a process for processing a consult question based on a client portrait in an embodiment of the present invention;
FIG. 6 is a structural schematic diagram of an apparatus for processing a consultative question based on a client portrait in an embodiment of the present invention;
FIG. 7 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making an invasive task, are within the scope of the present invention.
The method for processing the consult question based on the client portrait can be applied to the application environment shown in fig. 1, wherein the client communicates with the server. The client includes, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of a plurality of servers.
In an embodiment, as shown in FIG. 2, a method for processing a consult question based on a client portrait is provided, which is described by taking the application of the method to the server in FIG. 1 as an example, and includes the following steps:
s10, obtaining client consulting information, and performing semantic analysis on the client consulting information to obtain consulting keywords.
Specifically, when the entry of the consult work order is detected, consult content entered by a consultant in the consult work order is acquired to generate a first text, a call record of the customer in a certain time period is matched in a customer database according to the identity information of the consult customer in the consult work order, call voice in the call record is translated into a second text, and the second text is combined with the first text to generate the consult information of the customer. When the consult information is generated, the textual information contained in the consult information is traversed, information such as the syntactic structure of a sentence in the consult information and the word meaning of a word in the sentence is analyzed, and the consult keyword is extracted according to the syntactic structure and the word meaning in combination with the context. For example, a client complains about a bank for 3 times in a week, the telephone content from the client to the bank is translated into a first text, and the consulting keywords such as the complaint, blackout, indemnity, etc. are extracted through semantic analysis.
Understandably, semantic analysis means that a certain formal representation capable of reflecting the meaning of a sentence is deduced according to the syntactic structure of the sentence and the word meaning of each real word in the sentence, and natural language which can be understood by human is converted into formal language which can be understood by a computer.
S20, crawling a client level label representing client behavior information and basic information through network data;
understandably, the sources of the network data can be various social, service, life and other application software and related network platforms which are used by the client.
Specifically, various social, service, life and other application software and network data of related network platforms are crawled, and client-level tags representing client behavior information and basic information are obtained from the social, service, life and other application software. The first-level label of the client comprises basic information such as gender, age, occupation and education degree of the client and behavior information such as interest and shopping evaluation. The basic information and the behavior information of the client-level tag client can change along with the change of time, for example, the education level of the client can be updated at any time. Therefore, the first-level label of the client can be updated and supplemented regularly according to the needs.
And S30, inputting the consult keyword and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label.
Inputting the keywords obtained through semantic analysis and the client first-level tags obtained by crawling network data into a client portrait model, matching and classifying the keywords and the client first-level tags through a mapping relation formed in the client portrait model to obtain tags for representing client consult character and tags for representing client related information to generate a client portrait of the client, and obtaining the tags for representing the client consult character from the client portrait as a client second tag. The consulting personality of the client can be specifically classified into a cold static personality type, a benefit type, a sensitive type and a complaint type, and besides the cold static personality type, the benefit type, the sensitive type and the complaint type, the consulting personality of the client can be improved and updated in time according to actual implementation conditions, so that the consulting personality of the client can be classified more accurately.
In a specific embodiment, client consult information is obtained, the client consult information is subjected to semantic analysis to obtain a consult keyword, and the keyword of the client consult is clarified; further, a client level label representing client behavior information and basic information is crawled through network data, and more existing information data of a client are obtained; inputting the consulting keywords and the first-level client tag into a client portrait model for analysis, generating a client portrait, and obtaining a second-level client tag, wherein the second-level client tag obtained based on the first-level client tag is richer, and the second-level client tag obtained based on the keywords is more accurate. The consultative processing advice is generated by the client portrait and the client secondary label to enable consultative processing personnel to more comprehensively and quickly make processing judgment on consultative processing problems, so that the requirements of the target client work order can be more accurately grasped, the work order processing speed is increased, the work order processing time is shortened, the time for visiting the client and analyzing the problems by telephone is reduced, the client satisfaction degree is improved, and the labor is saved.
Optionally, in step S10, as shown in fig. 3, the method for acquiring the client consulting information includes the following steps:
s101, when the situation that the consulting work order is input is detected, the content of the consulting, which is input by the attending staff in the consulting work order, is acquired to generate a first text.
Specifically, when the system detects that a new work order is entered by an agent, the system detects that a new work order is created, sends a request for generating a first text after the entry of the consulting work order is completed and saved, acquires the content of the consulting of the client, which is entered by the agent in the consulting work order, and generates the content of the consulting into the first text.
S102, obtaining the call records of the client in the consulting work order in the client database within a period of time.
Specifically, the identity information of the consult client in the consult work order is acquired, the call records of the client within a period of time are matched in the client database according to the identity information of the client, and the client database is understandable and used for storing the historical call records and the call voice of the client. The period of time may be a preset period of time, such as a month or a quarter. The call records are records of telephone consultations for the customer about the problems of the service attitude, the service cost and the like of the customer service. Generally, the consult problem of a client can be communicated and solved through a plurality of calls, so that the obtaining of the call records in a certain period of time can better reflect the complaint and request of the client, and the obtaining of the call records in a certain period of time can also feed back the condition of the historical consult of the client.
S103, translating the call voice in the call record into a second text, and combining the second text with the first text to generate the client consulting information.
Specifically, after the call records of the client within a period of time are obtained, the call voices stored in the call records are sequentially obtained, and the obtained call voices are translated into the second text. And after the translation of the second text is finished, adding the second text into the first text, and combining the first text and the second text to generate the client consulting information. The client consulting information not only comprises the problems of the service attitude, the service cost and the like of the client for customer service in the current time and a certain time period, but also comprises the frequency and the times of the client consulting in a preset time period.
In a specific embodiment, when the presence of the consulting work order is detected, acquiring the content of the consulting, which is input by a consultant in the consulting work order, to generate a first text; acquiring a call record of a client in the consulting work order in a client database within a period of time; and translating the call voice in the call record into a second text, and combining the second text with the first text to generate the client consulting information. The client consulting information is more perfect by obtaining the voice call of the client and the work order of the client in a certain period of time.
Optionally, in step S30, the client secondary label characterizes the consult complaint of the client, and the consult complaint personality can be specifically classified into a cool and still personality, a benefit personality, a sensitive personality and a complaint personality.
Optionally, in step S30, the consultative can also be processed according to the client second-level tag to guide the consultative staff to deal with the problem.
Specifically, for example, the cool-still type indicates that the customer is cool and reasonable in financing and shopping without blindly consuming or purchasing financing services. When the client is in a cool and static type, the consultative staff is advised to consult the matters of the client by telephone without guiding the recommendation service blindly;
the profit type represents that the client has a loss in the aspect of financing and shopping. When the client is of interest type, the consultant is advised to consider that the client possibly wants to obtain corresponding compensation through consultant, and the telephone client carries out call return visit regularly after communication processing;
the sensitive type indicates that the information of the client in the aspect of financing and shopping is sensitive and is easily interfered by some information. When the client is sensitive, the consultative staff is advised to advise to consult and consult for multiple patience and multiple communication with the client when the client calls;
the complaint type indicates that the client is not cool enough in financing and shopping and usually likes consult. When the customer is of a complaint type, the customer needs to be careful and calmed, call return visit is carried out at regular time, and the customer is guided to reason financing and shopping.
In addition, the consulting personality of the client can be improved and updated in time according to the actual implementation situation, so that the consulting personality of the client can be classified more accurately.
Optionally, in step S10, as shown in fig. 4, the method for performing semantic analysis on the client consulting information to obtain the consultative keyword includes the following steps:
s104, automatically learning the client consulting information by using a shallow neural network model, representing words contained in the client consulting information as word vectors, and performing data preprocessing on the consulting information to obtain candidate consulting keywords.
Specifically, the shallow neural network model is utilized to automatically learn the client consult information, the occurrence condition of words in the client consult information is counted, the words are embedded into a high-dimensional space of 100-500, and the words are expressed in the form of word vectors in the high-dimensional space.
Furthermore, data preprocessing operations such as word segmentation, part-of-speech tagging, duplicate removal and stop word removal can be performed on the client consult information. And then obtaining a plurality of candidate consultations keywords by adopting the final participle of the accurate mode for the pre-processing client consultations information obtained through the data pre-processing operation. The final participle is generally divided into three modes, namely a full mode, a precise mode and a search engine mode: in the full mode, all words which can be formed into words in a sentence are scanned, so that the speed is very high, but ambiguity cannot be solved; the accurate mode is used for trying to cut the sentence most accurately, and is suitable for text analysis; and in the search engine mode, long words are segmented again on the basis of the accurate mode, the recall rate is improved, and the method is suitable for word segmentation of the search engine.
S105, obtaining a candidate consultative keyword vector according to the candidate consultative keyword, and performing cluster analysis on the candidate consultative keyword vector through a K-Means algorithm to obtain the consultative keyword.
Specifically, the candidate consulting keyword is traversed, a word vector corresponding to the candidate consulting keyword is obtained from the word vector and is represented as a candidate consulting keyword vector, clustering analysis is performed on the candidate consulting keyword vector through a K-Means algorithm to obtain a clustering center of each category, the Euclidean distance between words in a group and the clustering center in each category is calculated, descending order arrangement is performed according to the distance between the words in the group and the clustering center, the ranking result of the candidate consulting keyword vector is obtained, and the candidate consulting keyword corresponding to the candidate consulting keyword vectors ranked in the front is used as the consulting keyword.
In a specific embodiment, the client consult information is automatically learned by using a shallow neural network model, words contained in the client consult information are expressed as word vectors, and data preprocessing is performed on the consult information to obtain consult candidate keywords; acquiring a candidate consultative keyword vector according to the candidate consultative keyword, performing cluster analysis on the candidate consultative keyword vector through a K-Means algorithm to obtain consultative keywords, and acquiring consultative keywords from the client consultative information, so that the client consultative information is more accurate and short, and more favorable for accurate classification.
Optionally, in step S20, as shown in fig. 5, before the step of inputting the consultative keyword and the client-level tag into the client portrait model for analysis, the method further includes:
s201, obtaining the customer information characteristics of the sample, and generating a training set.
Specifically, sample client information features are obtained through data crawling, and a training set is generated through the obtained sample client information features. Wherein, the information characteristics comprise related information such as gender, age, occupation, education degree, interest, financing, shopping and the like.
S202, calculating the kini coefficient of each information feature to the training set, and selecting the information feature with the minimum kini coefficient to classify the training set to obtain the sub-training set to be processed.
Understandably, the kini coefficient of the training set reflects the probability that two samples, whose classes are not identical, are randomly drawn from the training set. The smaller the kini index is, the higher the purity of the data set is, and compared with information gain, information gain ratio and the like which are used as characteristic selection methods, the kini index omits logarithmic calculation and is smaller in calculation amount.
In the classification problem, it is assumed that the training set has k classes, and the probability of the k class is pKThen the expression of the k-th class for the kini coefficient of the training set is:
and sequentially calculating the basic damping coefficients of each information characteristic pair training set according to the expression formula of the basic damping coefficients, sequencing the basic damping coefficients according to a descending order, and selecting the information characteristic with the minimum basic damping coefficient as a classification node to classify the training sets to obtain the sub-training sets to be processed.
And S203, carrying out recursive calling on the to-be-processed sub training set to generate a client portrait model.
Specifically, the sub-training set to be processed is recursively called, that is, the remaining information features are called in step S202, and the client image model is also generated immediately when the recursive call is terminated when the remaining information features are less than or equal to zero.
Optionally, the generated customer representation model also needs to be validated for accuracy of its classification.
Optionally, the generated customer representation model may also be pruned.
Specifically, each classification node of the customer figure model is traversed, the node loss of each classification node is calculated, all to-be-processed sub-training sets of the customer figure model with the classification nodes as root nodes are traversed, and the loss of each to-be-processed sub-training set is calculated.
Understandably, the number of nodes of the customer portrait model is T, and the expression of the loss function is as follows:
Ca(T)＝C(T)+a|T|
where C (T) is the loss of the client representation model to the training set. a ≧ 0, a represents a parameter that balances training set loss C (T) and node number T, and Ca (T) represents the overall loss of the client portrait model. When a is larger, the smaller | T | is, the smaller the scale of the client representation model is; conversely, when a is smaller, | T | is larger, the size of the client representation model is larger.
For example, consider node t of the client representation model as a leaf node whose penalty is calculated:
Ca(t)＝C(t)+a|1|
and then calculating the loss of the sub-training set Tt to be processed when t is taken as a root node:
Ca(Tt)＝C(Tt)+a|Tt|
and when the node loss of a certain classification node is less than or equal to the loss of the to-be-processed sub-training set taking the classification node as the root node, pruning the classification node.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In one embodiment, the apparatus for processing the consultative problem based on the client portrait corresponds to the method for processing the consultative problem based on the client portrait in the above embodiment. As shown in FIG. 6, the apparatus for processing the consult question based on the client portrait comprises a keyword module, a first-level tag module and a second-level tag module. The functional modules are explained in detail as follows:
the key word module is used for acquiring client consult information and performing semantic analysis on the client consult information to obtain consult keywords;
the first-level label module is used for crawling a first-level label of the client representing the behavior information and the basic information of the client through network data;
and the secondary label module is used for inputting the consult keyword and the client primary label into a client portrait model for analysis to generate a client portrait and obtain a client secondary label.
Optionally, the keyword module, that is, the acquiring of the client consulting information, includes: a first text unit, a call log unit and a consult information unit.
The first text unit is used for acquiring the content of the consult recorded by the consultant in the consult work order to generate a first text when the consult work order is detected to be entered;
the call record unit is used for acquiring call records of the clients in the consult work order in a client database within a period of time;
and the consult information unit is used for translating the call voice in the call record into a second text, and generating client consult information by combining the second text with the first text.
Optionally, the secondary label module further includes: and a secondary label unit.
And the secondary label unit is used for representing the consult complaint character of the client, and the consult complaint character can be specifically classified into a cold-static character type, a benefit type, a sensitive type and a complaint type.
Optionally, the secondary label module further includes: consult and consult the suggestion unit.
The consult suggestion unit is used for generating consult treatment suggestion according to the client second-level tag to guide consult treatment personnel to treat the consult problem.
Optionally, the keyword module, that is, the method for performing semantic analysis on the client consulting information to obtain the consulting keyword includes:
the candidate consultative keyword unit is used for utilizing a shallow neural network model to automatically learn the client consultative information, representing words contained in the client consultative information as word vectors, and performing data preprocessing on the consultative information to obtain candidate consultative keywords;
and the consult key word unit is used for acquiring a consult key word vector according to the consult key word candidate, and performing cluster analysis on the consult key word vector through a K-Means algorithm to obtain a consult key word.
Optionally, before the second-level tag module, before inputting the consult keyword and the first-level tag of the client into the client portrait model for analysis, the method further includes:
the training set unit is used for acquiring the information characteristics of the sample clients and generating a training set;
the to-be-processed sub-training set unit is used for calculating the kini coefficient of each information characteristic to the training set, selecting the information characteristic with the minimum kini coefficient to divide the training set, and obtaining a to-be-processed sub-training set;
and the client portrait model unit is used for carrying out recursive calling on the to-be-processed sub training set to generate a client portrait model.
For specific limitations of the apparatus for processing the consultative problem based on the client portrait, reference may be made to the above limitations of the method for processing the consultative problem based on the client portrait, which are not described herein again. All modules in the device for processing the consult question based on the client image can be completely or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer readable instructions, and a database. The internal memory provides an environment for the operating system and execution of computer-readable instructions in the readable storage medium. The database of the computer device is used for storing data related to the method for processing the consult question based on the client portrait. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer readable instructions, when executed by a processor, implement a method for processing a consult question based on a client portrait. The readable storage media provided by the present embodiment include non-volatile readable storage media and volatile readable storage media.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a readable storage medium, an internal memory. The non-volatile storage medium stores an operating system and computer readable instructions. The internal memory provides an environment for the operating system and execution of computer-readable instructions in the readable storage medium. The network interface of the computer device is used for communicating with an external server through a network connection. The computer readable instructions, when executed by a processor, implement a method for processing a consult question based on a client portrait. The readable storage media provided by the present embodiment include non-volatile readable storage media and volatile readable storage media.
In one embodiment, a computer device is provided, comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, the processor when executing the computer readable instructions implementing the steps of:
acquiring client consult information, and performing semantic analysis on the client consult information to obtain consult key words;
crawling a client first-level label representing client behavior information and basic information through network data;
and inputting the consulting keyword and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label.
In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided, the readable storage media provided by the embodiments including non-volatile readable storage media and volatile readable storage media. The readable storage medium has stored thereon computer readable instructions which, when executed by one or more processors, perform the steps of:
acquiring client consult information, and performing semantic analysis on the client consult information to obtain consult key words;
crawling a client first-level label representing client behavior information and basic information through network data;
and inputting the consulting keyword and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label.
It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to computer readable instructions, which can be stored in a non-volatile readable storage medium or a volatile readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and simplicity of description, the foregoing functional units and modules are merely illustrated in terms of division, and in practical applications, the foregoing functional allocation may be performed by different functional units and modules as needed, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above described functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method for processing a consult question based on a client portrait is characterized by comprising the following steps:
acquiring client consult information, and performing semantic analysis on the client consult information to obtain consult keywords;
crawling a client first-level label representing client behavior information and basic information through network data;
and inputting the consult keywords and the client first-level label into a client portrait model for analysis, generating a client portrait and obtaining a client second-level label.
2. The method for processing the consultative problem based on the client portrait, as recited in claim 1, wherein the method for obtaining the client consultative information comprises:
when the consulting work order is detected to be entered, acquiring consulting content entered by a staff in the consulting work order to generate a first text;
acquiring a call record of a client in the consulting work order in a client database within a period of time;
and translating the call voice in the call record into a second text, and combining the second text with the first text to generate the client consulting information.
3. The method of claim 2, wherein the consulting question is processed based on a client portrait,
the client secondary label represents the consult complaint character of the client, and the consult complaint character can be specifically classified into a cold static character type, a benefit type, a sensitive type and a complaint type.
4. The method for processing the consultative problem based on the client portrait, as claimed in claim 3, wherein the consultative process advice guidance consult is generated to deal with the consultative problem according to the client second-level tag.
5. The method for processing the live action problem based on the client portrait, as claimed in claim 1, wherein the method for performing semantic analysis on the client live action information to obtain the live action keyword comprises the following steps:
utilizing a shallow neural network model to automatically learn the client consult information, representing words contained in the client consult information as word vectors, and carrying out data preprocessing on the consult information to obtain consult candidate keywords;
and obtaining a candidate consultative keyword vector according to the candidate consultative keyword, and performing cluster analysis on the candidate consultative keyword vector through a K-Means algorithm to obtain a consultative keyword.
6. The method of claim 1, wherein said inputting said consult keyword and said client-level tag into a client portrait model for analysis further comprises:
acquiring the information characteristics of sample clients to generate a training set;
calculating the kini coefficient of each information characteristic to the training set, and selecting the information characteristic with the minimum kini coefficient to divide the training set to obtain a to-be-processed sub-training set;
and carrying out recursive call on the to-be-processed sub training set to generate a client portrait model.
7. An apparatus for processing a consult question based on a client portrait, comprising:
the key word module is used for acquiring client consult information, performing semantic analysis on the client consult information and acquiring consult key words;
the first-level label module is used for crawling a first-level label of the client representing the behavior information and the basic information of the client through network data;
and the secondary label module is used for inputting the consult keyword and the client primary label into a client portrait model for analysis to generate a client portrait and obtain a client secondary label.
8. The apparatus for processing the consultative problem based on the client portrait, as recited in claim 7, wherein said obtaining the client consultative information comprises:
the first text unit is used for acquiring the content of the consult recorded by the consultant in the consult work order to generate a first text when the consult work order is detected to be entered;
the call record unit is used for acquiring call records of the clients in the consult work order in a client database within a period of time;
and the consult information unit is used for translating the call voice in the call record into a second text and combining the second text with the first text to generate the client consult information.
9. A computer apparatus comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor when executing the computer readable instructions implements the method for processing a problem of consulting based on a customer portrait according to any one of claims 1-6.
10. One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the method for processing the problem of consulting based on the customer representation as recited in any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110423425.3A CN113112282A (en) | 2021-04-20 | 2021-04-20 | Method, device, equipment and medium for processing consult problem based on client portrait |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110423425.3A CN113112282A (en) | 2021-04-20 | 2021-04-20 | Method, device, equipment and medium for processing consult problem based on client portrait |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113112282A true CN113112282A (en) | 2021-07-13 |
Family
ID=76718630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110423425.3A Pending CN113112282A (en) | 2021-04-20 | 2021-04-20 | Method, device, equipment and medium for processing consult problem based on client portrait |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113112282A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113822062A (en) * | 2021-09-22 | 2021-12-21 | 未鲲(上海)科技服务有限公司 | Text data processing method, device and equipment and readable storage medium |
| CN114121039A (en) * | 2021-11-19 | 2022-03-01 | 北京京东振世信息技术有限公司 | Sound processing method, device, equipment and storage medium |
| CN117556802A (en) * | 2024-01-12 | 2024-02-13 | 碳丝路文化传播(成都)有限公司 | User portrait method, device, equipment and medium based on large language model |
| CN117610891A (en) * | 2024-01-22 | 2024-02-27 | 湖南小翅科技有限公司 | Flexible work order and risk control system based on big data |
| CN118503546A (en) * | 2024-07-18 | 2024-08-16 | 广州博今网络技术有限公司 | Form data pushing method and system based on associated objects |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222333A (en) * | 2019-05-20 | 2019-09-10 | 平安普惠企业管理有限公司 | A kind of voice interactive method, device and relevant device |
| CN111210201A (en) * | 2020-01-02 | 2020-05-29 | 平安科技（深圳）有限公司 | Occupational label establishing method and device, electronic equipment and storage medium |
| CN111274380A (en) * | 2020-01-16 | 2020-06-12 | 平安银行股份有限公司 | Consultation complaint information processing method based on big data and related device |
-
2021
- 2021-04-20 CN CN202110423425.3A patent/CN113112282A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222333A (en) * | 2019-05-20 | 2019-09-10 | 平安普惠企业管理有限公司 | A kind of voice interactive method, device and relevant device |
| CN111210201A (en) * | 2020-01-02 | 2020-05-29 | 平安科技（深圳）有限公司 | Occupational label establishing method and device, electronic equipment and storage medium |
| CN111274380A (en) * | 2020-01-16 | 2020-06-12 | 平安银行股份有限公司 | Consultation complaint information processing method based on big data and related device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113822062A (en) * | 2021-09-22 | 2021-12-21 | 未鲲(上海)科技服务有限公司 | Text data processing method, device and equipment and readable storage medium |
| CN114121039A (en) * | 2021-11-19 | 2022-03-01 | 北京京东振世信息技术有限公司 | Sound processing method, device, equipment and storage medium |
| CN117556802A (en) * | 2024-01-12 | 2024-02-13 | 碳丝路文化传播(成都)有限公司 | User portrait method, device, equipment and medium based on large language model |
| CN117556802B (en) * | 2024-01-12 | 2024-04-05 | 碳丝路文化传播(成都)有限公司 | User portrait method, device, equipment and medium based on large language model |
| CN117610891A (en) * | 2024-01-22 | 2024-02-27 | 湖南小翅科技有限公司 | Flexible work order and risk control system based on big data |
| CN117610891B (en) * | 2024-01-22 | 2024-04-02 | 湖南小翅科技有限公司 | Flexible work order and risk control system based on big data |
| CN118503546A (en) * | 2024-07-18 | 2024-08-16 | 广州博今网络技术有限公司 | Form data pushing method and system based on associated objects |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12039280B2 (en) | Multi-turn dialogue response generation with persona modeling | |
| US11907274B2 (en) | Hyper-graph learner for natural language comprehension | |
| CN113112282A (en) | Method, device, equipment and medium for processing consult problem based on client portrait | |
| CN110222167B (en) | Method and system for acquiring target standard information | |
| CN108536800B (en) | Text classification method, system, computer device and storage medium | |
| CN110738545A (en) | Product recommendation method and device based on user intention identification, computer equipment and storage medium | |
| WO2021139343A1 (en) | Data analysis method and apparatus based on natural language processing, and computer device | |
| US11765267B2 (en) | Tool for annotating and reviewing audio conversations | |
| CN112380331A (en) | Information pushing method and device | |
| US20160283876A1 (en) | System and method for providing automomous contextual information life cycle management | |
| CN110377733A (en) | A kind of text based Emotion identification method, terminal device and medium | |
| CN112256863A (en) | Method and device for determining corpus intentions and electronic equipment | |
| CN113919437A (en) | Method, device, equipment and storage medium for generating client portrait | |
| CN114399396A (en) | Insurance product recommendation method and device, computer equipment and storage medium | |
| WO2019191817A1 (en) | A system and method for generating documents | |
| CN112749238A (en) | Search ranking method and device, electronic equipment and computer-readable storage medium | |
| CN112597292B (en) | Question reply recommendation method, device, computer equipment and storage medium | |
| US20240184838A1 (en) | Diversified recommendation method for news based on graph neural network and its device | |
| CN113870998A (en) | Interrogation method, device, electronic equipment and storage medium | |
| CN112860878A (en) | Service data recommendation method, storage medium and equipment | |
| CN116525093A (en) | Prediction method, device, equipment and storage medium for session ending | |
| CN113961811A (en) | Conversational recommendation method, device, equipment and medium based on event map | |
| CN114417045A (en) | Insurance case spot inspection method, system, equipment and storage medium based on neural network | |
| Kasper et al. | User profile acquisition: A comprehensive framework to support personal information agents | |
| WO2019236338A1 (en) | Computerized relevance scoring engine for identifying potential investors for a new business entity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |