CN117633626A - Model updating method, device and computer equipment - Google Patents
Model updating method, device and computer equipment Download PDFInfo
- Publication number
- CN117633626A CN117633626A CN202311651188.1A CN202311651188A CN117633626A CN 117633626 A CN117633626 A CN 117633626A CN 202311651188 A CN202311651188 A CN 202311651188A CN 117633626 A CN117633626 A CN 117633626A
- Authority
- CN
- China
- Prior art keywords
- target
- classification
- data
- service
- model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present application relates to the field of big data processing technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for updating a model. The model updating method comprises the following steps: responding to the processing instruction, and extracting target characteristic data from target service data carried by the processing instruction; inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data; according to the classification result, matching the target business response rule corresponding to the target business data from at least one preset business response rule; and acquiring a target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result. Through the arrangement, the accuracy of the classification result is improved, and the finally obtained updated model can be closest to the actual classification requirement, so that the classification accuracy is improved, and the subsequent data analysis is facilitated.
Description
Technical Field
The present invention relates to the field of big data processing technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for updating a model.
Background
With the rapid development of computer technology, internet systems have provided a variety of service systems in order to meet the demands of users for processing various transactions, with concurrent increases in service data for different services.
In order to better understand the service processing requirements of the same cluster or the same service object, the service data needs to be classified correspondingly for subsequent analysis, and in general, the classification of a large amount of service data is performed by service personnel according to experience, which integrates a plurality of subjective judgment elements, so that the deviation of the finally obtained classification result is larger; if the model is adopted to classify the service data, a large amount of resources are consumed by collecting a large amount of sample data required by training the model, and the accuracy of classifying the model obtained by training is not high.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a model updating method, apparatus, computer device, storage medium, and computer program product that can improve the accuracy of data classification of a model.
In a first aspect, the present application provides a method for updating a model, including:
responding to a processing instruction, and extracting target characteristic data from target service data carried by the processing instruction; the target feature data carries at least one type of interaction data;
Inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data;
according to the classification result, matching a target service response rule corresponding to the target service data from at least one preset service response rule;
acquiring a target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result;
the step of obtaining the target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result, includes:
obtaining a target sub-feedback result from at least one feedback terminal corresponding to a pre-designated service provider and/or at least one feedback terminal corresponding to a service demand party;
determining the target feedback result according to each target sub-feedback result;
matching the target feedback result with a corresponding target classification label from at least one classification label; the classification label is used for identifying a classification result;
storing the target business data and the target classification labels to the sample database correspondingly to obtain a new sample database;
And updating the classification model by adopting a new sample database to obtain an updated classification model.
In one embodiment, the extracting, in response to a processing instruction, target feature data from target service data carried by the processing instruction includes:
responding to the processing instruction, and performing data cleaning processing on target service data carried by the processing instruction;
and extracting the target characteristic data from the target business data after the cleaning treatment.
In one embodiment, the training process of the classification model includes:
collecting at least one sample service data to establish a sample database; the sample business data is marked with a sample classification label in advance;
dividing the sample business data into a training set and a testing set according to a preset proportion;
inputting the training set into an initial classification model for training to obtain a trained classification model;
inputting the test set into a trained classification model to obtain sample classification results corresponding to the sample service data in the test set;
and when the coincidence degree of the sample classification result and the sample classification label corresponding to each sample service data in the test set reaches a preset requirement, determining that the classification model training is completed.
In one embodiment, after updating the classification model with the new sample database to obtain an updated classification model, the method further comprises:
and responding to the new processing instruction by adopting the updated classification model to obtain a classification result corresponding to the new processing instruction.
In one embodiment, the target feedback result includes a target feedback value;
at least one business response rule is preset with an arrangement sequence, and the business response rule corresponds to the classification labels one by one;
the matching from at least one sort label to a corresponding target sort label according to the target feedback result comprises:
matching a target value range corresponding to the target feedback value from at least one value range according to the target feedback value; the numerical ranges are in one-to-one correspondence with the adjustment rules;
according to the target numerical value range, matching an adjustment rule corresponding to the target feedback result;
according to the adjustment rule and the arrangement sequence, determining an adjustment service response rule corresponding to the target feedback result;
and taking the classification label corresponding to the adjustment service response rule as a target classification label corresponding to the target feedback result.
In one embodiment, the adjustment rule includes a first preset number of business response rules arranged before or after the reference position as the adjustment business response rule.
The determining, according to the target adjustment rule and the arrangement order, an adjustment service response rule corresponding to the target feedback result includes:
according to the arrangement sequence, taking the arrangement position of the target service response rule as a reference position;
and taking a first preset numerical business response rule arranged before or after the reference position as an adjustment business response rule corresponding to the target feedback result.
In a second aspect, the present application further provides a model updating apparatus, including:
the extraction module is used for responding to the processing instruction and extracting target characteristic data from target service data carried by the processing instruction;
the classification module is used for inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data;
the matching module is used for matching the target service response rule corresponding to the target service data from at least one preset service response rule according to the classification result;
The updating module is used for acquiring a target feedback result corresponding to the target service response rule and updating the classification model based on the target characteristic data and the target feedback result; and the target feedback result is determined according to the target sub-feedback result sent by at least one feedback terminal.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the model updating method according to any of the above embodiments.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the model updating method described in any of the above embodiments.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the model updating method according to any of the above embodiments.
According to the model updating method, device, computer equipment, storage medium and computer program product, when the processing instruction is received, different types of interaction data carried by the processing instruction can be extracted, the target feature data is further extracted and input into the classification model, a corresponding classification result is obtained, the target feature data contains a large number of different types of data features, the data island is broken, the accuracy of the classification result is improved, the target business response rule applicable to the business demander corresponding to the current target business data can be determined according to the classification result, at least one target sub-feedback result is received from at least one feedback terminal, and finally the classification model is updated according to at least one target sub-feedback result, so that the classification model can be updated and adjusted in real time after data classification is carried out each time, the finally obtained updated model can be closest to the actual classification requirement, the classification accuracy is improved, and the subsequent data analysis is facilitated.
Drawings
FIG. 1 is a diagram of an application environment for a model update method in one embodiment;
FIG. 2 is a flow diagram of a model update method in one embodiment;
FIG. 3 is a flow chart of a model update method according to yet another embodiment;
FIG. 4 is a flow chart of a model update method according to another embodiment;
FIG. 5 is a flow chart of a model update method in yet another embodiment;
FIG. 6 is a block diagram of a model updating device in one embodiment;
fig. 7 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. The model updating method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 over a network and the server 104 corresponds to the initial component.
For example, the model updating method is applied to the terminal 102, and when the terminal 102 receives a processing instruction, the terminal extracts target feature data from target service data carried by the processing instruction; inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data; the terminal 102 then obtains at least one preset service response rule from the data storage system of the server 104, and matches the target service response rule corresponding to the target service data from the at least one service response rule according to the classification result; finally, a target feedback result corresponding to the target service response rule is obtained, and a classification model is updated based on the target characteristic data and the target feedback result; the target feedback result is determined according to a target sub-feedback result sent by at least one service terminal, wherein the terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers. The first server 104 and the second server 106 may be implemented as separate servers or as a server cluster composed of a plurality of servers. The terminal 102 and the first server 104 and the second server 106 may be directly or indirectly connected through wired or wireless communication means, for example, through a network connection.
For another example, the service processing method is applied to the server 104, when the terminal 102 receives the processing instruction, the processing instruction may be sent to the server 104, and the server 104 extracts the target feature data from the target service data carried by the processing instruction; inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data; the server 104 then obtains at least one preset service response rule from the data storage system, and matches the target service response rule corresponding to the target service data from the at least one service response rule according to the classification result; finally, a target feedback result corresponding to the target service response rule is obtained, and a classification model is updated based on the target characteristic data and the target feedback result; and determining a target feedback result according to the target sub-feedback result sent by at least one service terminal. It will be appreciated that the data storage system may be a stand-alone storage device, or the data storage system may be located on a server, or the data storage system may be located on another terminal.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In one embodiment, a model updating method is provided, and the embodiment is exemplified by the application of the model updating method to a terminal, and it can be understood that the method can also be applied to a server, and can also be applied to a system comprising the terminal and the server, and implemented through interaction between the terminal and the server. As shown in fig. 2, the model updating method includes:
Step 202, responding to a processing instruction, and extracting target characteristic data from target service data carried by the processing instruction; the target characteristic data carries at least one type of interaction data.
The processing instructions may refer to instructions for relevant business responses to the business demander.
The processing instruction can be sent by a worker of the service provider through a man-machine interaction interface of the terminal, the man-machine interaction interface of the terminal can specifically display a platform interface appointed by the service provider, and the worker of the service demand party sends the processing instruction in a mode of clicking an area corresponding to any service demand party on the appointed platform interface; or, the processing instruction may be automatically generated when the terminal receives a related trigger instruction sent by any service requiring party; the related trigger instruction may be sent by the service demander when the platform interface designated by the service provider performs the registration action, or may be sent by the service demander when the platform interface designated by the service provider clicks a specific area, where the specific area may be a push button for subscribing the related information of the service provider, for example.
The target business data may be business-requiring party related interaction data.
The target service data may include different types of interaction data corresponding to the service demand party, for example, the target service data may include information such as quantity information of interaction products and specification information of interaction products generated when the service demand party adopts an e-commerce system to perform service processing; when the business demand party adopts the multiparty interaction support system to interact with other business demand parties, the generated information such as identity information of the interaction parties, quantity information of interaction resources and the like is adopted.
The target feature data refers to data for describing attributes or features of different types of interaction data corresponding to the business demander.
And 204, inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data.
The initial classification model may be any one of a K-nearest neighbor classification learning model, a support vector machine classification learning model, a neural network classification learning model, or an ensemble learning model.
The classification result may refer to a result of classifying the target feature data according to specific data contents of the target feature data. The classification result is used to characterize the interaction characteristics of the business demander. When the target feature data comprises information such as the quantity information of the interactive products, the specification information of the interactive products and the like generated when the business demand side carries out business processing by adopting an electronic commerce system, the classification result can be used for representing the interactive demand of the business demand side during product interaction; when the target feature data contains information such as identity information of both interaction parties and quantity information of interaction resources generated when the service demand party interacts with other service demand parties by adopting the multiparty interaction support system, the classification result can be used for representing the interaction demand of the service demand party during multiparty interaction.
And 206, matching the target business response rule corresponding to the target business data from at least one preset business response rule according to the classification result.
The service response rule refers to a series of regulations and standards formulated by a service provider for a service request of a service demander or for an invisible service demander, so as to ensure timely and effective response to the service requirement of the service demander and introduction and pushing of related services which can be provided by the service provider to the invisible service demander.
The service response rules are in one-to-one correspondence with various classification results which can be output by the trained classification model, and when the terminal outputs the classification result corresponding to the target service data according to the target feature data, the terminal can be matched with the target service response rules corresponding to the target service data from at least one service response rule according to the classification result corresponding to the target service data.
Step 208, obtaining a target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result; and determining a target feedback result according to the target sub-feedback result sent by at least one feedback terminal.
The target feedback result refers to feedback comments or evaluations provided by the service quality and the processing result of the service provider after the service response is performed on the service demand party by adopting the target service response rule. The target feedback result may be sent by a worker of the service demand party through at least one feedback terminal, or may be sent by a worker pre-designated by the service provider through at least one preset feedback terminal according to the actual feedback of the service demand party.
Further, the terminal may pre-allocate a corresponding weight value to at least one feedback terminal capable of sending the target sub-feedback result, and when the terminal receives the target sub-feedback result from the at least one feedback terminal, obtain a final target feedback result according to the weight value corresponding to the feedback terminal sending each target sub-feedback result and the target sub-feedback result.
The target feedback result can reflect satisfaction information, complaint information, advice information and the like of the business demander. By collecting and analyzing the target feedback result, the terminal can obtain the satisfaction degree of the service demand party on the target service response rule, so that the classification and response of the target service data of other subsequent service demand parties are continuously optimized.
According to the login state processing method, when the terminal receives the processing instruction, the terminal can extract different types of interaction data carried by the processing instruction, further can extract target feature data from the processing instruction and input the target feature data into the classification model to obtain a corresponding classification result, the target feature data comprises a large number of different types of data features, the data island is broken, the accuracy of the classification result is improved, the target business response rule applicable to a business demander corresponding to current target business data can be determined according to the classification result, at least one target sub-feedback result is received from at least one feedback terminal, and finally the classification model is updated according to at least one target sub-feedback result, so that after data classification is carried out each time, the classification model can be updated and adjusted in real time, the finally obtained updated model can be closest to the actual classification requirement, classification accuracy is improved, and subsequent data analysis is facilitated.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. As shown in fig. 3, in some alternative embodiments, step 202 includes:
Step 2022, responding to the processing instruction, and performing data cleaning processing on the target service data carried by the processing instruction;
step 2024, extracting target feature data from the target service data after the cleaning process.
The data cleaning refers to a process of identifying, correcting, deleting or modifying inaccurate, incomplete, repeated or inapplicable data content of target service data, and aims to ensure the quality of the target service data so as to extract accurate target feature data.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the training process of the classification model includes:
collecting at least one sample service data to establish a sample database; sample business data is marked with sample classification labels in advance;
dividing sample business data into a training set and a testing set according to a preset proportion;
inputting the training set into an initial classification model for training to obtain a trained classification model;
inputting the test set into a trained classification model to obtain sample classification results corresponding to the sample service data in the test set;
And when the coincidence degree of the sample classification result and the sample classification label corresponding to each sample service data in the test set reaches a preset requirement, determining that the classification model training is completed.
Sample business data refers to a plurality of known business demander-related interaction data.
The preset requirement may be that the coincidence degree of the sample classification result and the sample classification label corresponding to each sample service data in the test set reaches the prediction accuracy threshold.
The sample classification label may be at least one of letters, characters or numbers, and is used for uniquely identifying a classification result corresponding to the sample service data, and the data classification result corresponding to the sample service data may be marked on the sample service data in advance by a staff of the service provider, where a mapping relationship between the sample classification label and multiple classification results is pre-stored in the terminal in this embodiment.
In this embodiment, the terminal may randomly divide all sample service data in the sample database according to a ratio of 7:3, take 70% of the sample service data as a training set, take 30% of the sample service data as a test set, input the training set into an initial classification model to obtain a trained classification model, input the test set into the trained classification model to obtain classification results corresponding to each sample service data in the test set, compare the classification results corresponding to each sample service data in the test set with sample classification labels corresponding to each sample service data in the test set, calculate the coincidence degree corresponding to each sample service data in the test set, and determine whether the coincidence degree reaches a prediction accuracy threshold, if so, obtain a classification model which is finally trained, and if not, re-train the classification model by using the test set until the coincidence degree corresponding to the classification model reaches the prediction accuracy threshold.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. As shown in fig. 4, in some alternative embodiments, step 208 includes:
step 2082, matching the target classification label corresponding to the at least one classification label according to the target feedback result;
step 2084, storing the target business data and the target classification label to a sample database correspondingly to obtain a new sample database;
and step 2086, updating the classification model by adopting a new sample database to obtain an updated classification model.
The classification label can be at least one of letters, characters or numbers, and is used for uniquely identifying the classification result, and the terminal in this embodiment stores the mapping relation between the classification label and various classification results in advance.
As an example, when the terminal obtains the target sub-feedback result from at least one feedback terminal, the target sub-feedback result may further include, for example, a recommended sub-service response rule sent by a staff of the service demander through the feedback terminal, or a recommended sub-service response rule sent by a staff pre-designated by the service provider through the feedback terminal, where the terminal determines a final recommended service response rule according to the recommended sub-service response rule corresponding to the at least one target sub-feedback result, and takes a classification tag corresponding to the recommended service response rule as a target classification tag.
The determining the final recommended service response rule according to the recommended sub-service response rule corresponding to the at least one target sub-feedback result may be to directly use the recommended sub-service response rule with the largest occurrence number as the recommended service response rule.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the model updating method further comprises:
and responding to the new processing instruction by adopting the updated classification model to obtain a classification result corresponding to the new processing instruction.
The classification model in this embodiment is updated and adjusted in real time according to the target feedback result corresponding to the target service response rule, when the terminal receives the new processing instruction, the terminal can extract new target feature data from the target service data carried by the new processing instruction, and input the new target feature data into the classification model after updating and adjusting to obtain the classification result corresponding to the new processing instruction, so that the classification result corresponding to the new processing instruction can better meet the actual classification requirement, and correspondingly, the target service response rule corresponding to the classification result can better meet the actual service requirement.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, as shown in FIG. 5, the target feedback result comprises a target feedback value;
step 2082 includes:
step 20822, matching the target feedback value to a target value range corresponding to the target feedback value from at least one value range according to the target feedback value; the numerical value ranges are in one-to-one correspondence with the adjustment rules;
step 20824, matching to a target adjustment rule corresponding to the target feedback result according to the target numerical range;
step 20826, determining an adjustment business response rule corresponding to the target feedback result according to the target adjustment rule and the arrangement sequence;
step 20828, using the classification label corresponding to the adjustment business response rule as the target classification label corresponding to the target feedback result.
In this embodiment, the staff of the service demand party may score the target service response rule through the feedback terminal, or the staff pre-designated by the service provider may score the target service response rule through the feedback terminal according to the actual feedback of the service demand party.
After receiving the score sent by at least one feedback terminal, the terminal can directly take the average number as a target feedback value.
The adjustment rule refers to a preset number of service response rules arranged before or after the reference position as an adjustment service response rule, and in this embodiment, according to the arrangement sequence of each service response rule, the service response rule corresponding to another arrangement position near the reference position may be determined by taking the arrangement position where the target service response rule is located as the reference position.
As an example, the adjustment rule corresponding to the target feedback result may be a previous business response rule for determining the reference position in the arrangement order as the adjustment business response rule.
The terminal takes the previous service response rule of the target service response rule as the adjustment service response rule according to the arrangement sequence.
According to the model updating method, when the terminal receives the processing instruction, the terminal can extract different types of interaction data carried by the processing instruction, further can extract target feature data from the processing instruction and input the target feature data into the classification model to obtain a corresponding classification result, the target feature data comprises a large number of different types of data features, the data island is broken, the accuracy of the classification result is improved, the target business response rule applicable to the business demander corresponding to the current target business data can be determined according to the classification result, at least one target sub-feedback result is received from at least one feedback terminal, finally, the adjustment business response rule corresponding to the adjustment processing instruction is determined according to the at least one target sub-feedback result, and the classification model is updated according to the adjustment business response rule, so that the finally obtained update model can be updated and adjusted in real time after data classification is carried out each time, the classification model can be closest to actual classification requirements, classification accuracy is improved, and follow-up data analysis is facilitated.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a model updating device for realizing the model updating method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in one or more embodiments of the model updating device provided below may refer to the limitation of the model updating method hereinabove, and will not be repeated here.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In one embodiment, as shown in fig. 6, there is provided a model updating apparatus 500, including: an extraction module 602, a classification module 604, a matching module 606, and an update module 608, wherein:
the extracting module 602 is configured to respond to the processing instruction, and extract target feature data from target service data carried by the processing instruction; the target feature data carries at least one type of interaction data;
the classification module 604 is configured to input the target feature data into a pre-trained classification model, so as to obtain a classification result corresponding to the target service data;
the matching module 606 is configured to match, according to the classification result, a target service response rule corresponding to the target service data from at least one preset service response rule;
the updating module 608 is configured to obtain a target feedback result corresponding to the target service response rule, and update the classification model based on the target feature data and the target feedback result; and determining a target feedback result according to the target sub-feedback result sent by at least one feedback terminal.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the extraction module 602 is further configured to:
Responding to the processing instruction, and performing data cleaning processing on target service data carried by the processing instruction;
and extracting target characteristic data from the target service data after the cleaning treatment.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the training process of the classification model includes:
collecting at least one sample service data to establish a sample database; sample business data is marked with sample classification labels in advance;
dividing sample business data into a training set and a testing set according to a preset proportion;
inputting the training set into an initial classification model for training to obtain a trained classification model;
inputting the test set into a trained classification model to obtain sample classification results corresponding to the sample service data in the test set;
and when the coincidence degree of the sample classification result and the sample classification label corresponding to each sample service data in the test set reaches a preset requirement, determining that the classification model training is completed.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the update module 608 is further configured to:
Matching the target classification label corresponding to the target classification label from at least one classification label according to the target feedback result;
storing the target business data and the target classification labels to a sample database correspondingly to obtain a new sample database;
and updating the classification model by adopting a new sample database to obtain an updated classification model.
In some alternative embodiments, the update module 608 is further configured to:
and responding to the new processing instruction by adopting the updated classification model to obtain a classification result corresponding to the new processing instruction.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In some alternative embodiments, the target feedback result comprises a target feedback value;
at least one business response rule is preset with an arrangement sequence, and the business response rules are in one-to-one correspondence with the classification labels;
the update module 608 is further configured to:
matching a target value range corresponding to the target feedback value from at least one value range according to the target feedback value; the numerical value ranges are in one-to-one correspondence with the adjustment rules;
matching an adjustment rule corresponding to the target feedback result according to the target numerical value range;
According to the adjustment rules and the arrangement sequence, determining an adjustment business response rule corresponding to the target feedback result;
and taking the classification label corresponding to the adjustment service response rule as a target classification label corresponding to the target feedback result.
The respective modules in the above-described model updating apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a model update method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the model updating method in any of the above embodiments.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the model updating method in any of the embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.
Claims (10)
1. A method of updating a model, comprising:
responding to a processing instruction, and extracting target characteristic data from target service data carried by the processing instruction; the target feature data carries at least one type of interaction data;
inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data; the classification model is obtained by training a sample database;
According to the classification result, matching a target service response rule corresponding to the target service data from at least one preset service response rule;
acquiring a target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result;
the step of obtaining the target feedback result corresponding to the target service response rule, and updating the classification model based on the target characteristic data and the target feedback result, includes:
obtaining a target sub-feedback result from at least one feedback terminal corresponding to a pre-designated service provider and/or at least one feedback terminal corresponding to a service demand party;
determining the target feedback result according to each target sub-feedback result;
matching the target feedback result with a corresponding target classification label from at least one classification label; the classification label is used for identifying a classification result;
storing the target business data and the target classification labels to the sample database correspondingly to obtain a new sample database;
and updating the classification model by adopting a new sample database to obtain an updated classification model.
2. The method of claim 1, wherein the extracting, in response to a processing instruction, target feature data from target service data carried by the processing instruction comprises:
responding to the processing instruction, and performing data cleaning processing on target service data carried by the processing instruction;
and extracting the target characteristic data from the target business data after the cleaning treatment.
3. The method of claim 1, wherein the training process of the classification model comprises:
collecting at least one sample service data and establishing the sample database; the sample business data is marked with a sample classification label in advance;
dividing the sample business data into a training set and a testing set according to a preset proportion;
inputting the training set into an initial classification model for training to obtain a trained classification model;
inputting the test set into a trained classification model to obtain sample classification results corresponding to the sample service data in the test set;
and when the coincidence degree of the sample classification result and the sample classification label corresponding to each sample service data in the test set reaches a preset requirement, determining that the classification model training is completed.
4. A method according to claim 3, wherein after said updating of said classification model with a new sample database resulting in an updated classification model, the method further comprises:
and responding to the new processing instruction by adopting the updated classification model to obtain a classification result corresponding to the new processing instruction.
5. A method according to claim 3, wherein the target feedback result comprises a target feedback value;
at least one business response rule is preset with an arrangement sequence, and the business response rule corresponds to the classification labels one by one;
the matching from at least one sort label to a corresponding target sort label according to the target feedback result comprises:
matching a target value range corresponding to the target feedback value from at least one value range according to the target feedback value; the numerical ranges are in one-to-one correspondence with the adjustment rules;
according to the target numerical value range, matching to a target adjustment rule corresponding to the target feedback result;
according to the target adjustment rule and the arrangement sequence, determining an adjustment service response rule corresponding to the target feedback result;
And taking the classification label corresponding to the adjustment service response rule as a target classification label corresponding to the target feedback result.
6. A method according to claim 3, wherein the adjustment rule includes, as the adjustment business response rule, a first preset number of business response rules arranged before or after the reference position;
the determining, according to the target adjustment rule and the arrangement order, an adjustment service response rule corresponding to the target feedback result includes:
according to the arrangement sequence, taking the arrangement position of the target service response rule as a reference position;
and taking a first preset numerical business response rule arranged before or after the reference position as an adjustment business response rule corresponding to the target feedback result.
7. A model updating apparatus, characterized by comprising:
the extraction module is used for responding to the processing instruction and extracting target characteristic data from target service data carried by the processing instruction; the target feature data carries at least one type of interaction data;
the classification module is used for inputting the target characteristic data into a pre-trained classification model to obtain a classification result corresponding to the target service data; the classification model is obtained by training a sample database;
The matching module is used for matching the target service response rule corresponding to the target service data from at least one preset service response rule according to the classification result;
the updating module is used for acquiring a target feedback result corresponding to the target service response rule and updating the classification model based on the target characteristic data and the target feedback result;
the update module is further configured to: obtaining a target sub-feedback result from at least one feedback terminal corresponding to a pre-designated service provider and/or at least one feedback terminal corresponding to a service demand party;
determining the target feedback result according to each target sub-feedback result;
matching the target feedback result with a corresponding target classification label from at least one classification label; the classification label is used for identifying a classification result;
storing the target business data and the target classification labels to the sample database correspondingly to obtain a new sample database;
and updating the classification model by adopting a new sample database to obtain an updated classification model.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the model updating method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the model updating method of any of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the model updating method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311651188.1A CN117633626A (en) | 2023-12-04 | 2023-12-04 | Model updating method, device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311651188.1A CN117633626A (en) | 2023-12-04 | 2023-12-04 | Model updating method, device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117633626A true CN117633626A (en) | 2024-03-01 |
Family
ID=90019617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311651188.1A Pending CN117633626A (en) | 2023-12-04 | 2023-12-04 | Model updating method, device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117633626A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119203244A (en) * | 2024-11-25 | 2024-12-27 | 众合云科信息技术集团有限公司 | A governance method for large model data leakage |
| WO2025236515A1 (en) * | 2024-05-16 | 2025-11-20 | 鹏城实验室 | Dynamic model update method and device, and storage medium and computer program product |
-
2023
- 2023-12-04 CN CN202311651188.1A patent/CN117633626A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2025236515A1 (en) * | 2024-05-16 | 2025-11-20 | 鹏城实验室 | Dynamic model update method and device, and storage medium and computer program product |
| CN119203244A (en) * | 2024-11-25 | 2024-12-27 | 众合云科信息技术集团有限公司 | A governance method for large model data leakage |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325353B (en) | Training data set supplier selection method, device, equipment and storage medium | |
| CN110070391B (en) | Data processing method and device, computer readable medium and electronic equipment | |
| CN109582876B (en) | Tourist industry user portrait construction method and device and computer equipment | |
| US10565525B2 (en) | Collaborative filtering method, apparatus, server and storage medium in combination with time factor | |
| CN110598845A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| US20230342797A1 (en) | Object processing method based on time and value factors | |
| CN117633626A (en) | Model updating method, device and computer equipment | |
| CN112070310A (en) | Loss user prediction method and device based on artificial intelligence and electronic equipment | |
| CN112950218A (en) | Business risk assessment method and device, computer equipment and storage medium | |
| CN114722281B (en) | Training course configuration method and device based on user portrait and user course selection behavior | |
| CN115730125A (en) | Object identification method and device, computer equipment and storage medium | |
| CN115439192A (en) | Medical commodity information pushing method and device, storage medium and computer equipment | |
| CN117194779A (en) | Marketing system optimization methods, devices and equipment based on artificial intelligence | |
| CN110598084A (en) | Object sorting method, commodity sorting device and electronic equipment | |
| CN115758271A (en) | Data processing method, device, computer equipment and storage medium | |
| CN115630221A (en) | Terminal application interface display data processing method and device and computer equipment | |
| US11295325B2 (en) | Benefit surrender prediction | |
| CN110942345B (en) | Seed user selection method, device, equipment and storage medium | |
| CN112580916B (en) | Data evaluation method, device, computer equipment and storage medium | |
| CN113434432A (en) | Performance test method, device, equipment and medium for recommendation platform | |
| CN119271777A (en) | Intelligent reply method, device, equipment and medium based on natural language processing | |
| CN108510071B (en) | Data feature extraction method and device and computer readable storage medium | |
| CN117057886A (en) | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium | |
| US20220253690A1 (en) | Machine-learning systems for simulating collaborative behavior by interacting users within a group | |
| KR20140073624A (en) | Methods for competency assessment of corporation for global business |
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 |