Work order data processing method and device, computer equipment, readable storage medium and program product
By extracting the Q&A pair of work ticket data and determining the target standard Q&A pair with historical data, the problem of low accuracy and efficiency of work ticket data processing in the prior art is solved, and higher accuracy and response speed are achieved.
Patent Information
- Application Number
- CN202411724230.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-03-14
AI Technical Summary
In the prior art, fine-tuning based on large language models has problems of knowledge updating difficulties, and retrieval-based question-and-answer systems have problems of fragmentation of knowledge, long response time and model hallucination risks, resulting in low accuracy and efficiency of work ticket data processing.
By obtaining the pending work order data, extracting the Q&A pair, and determining the first standard Q&A pair based on the pending work order and the historical Q&A pair; determining the second standard Q&A pair based on the pending work order data and historical work order data; combining the first and second standard Q&A pair, determining the target standard Q&A pair, and then updating the historical data to improve accuracy.
By expanding the scope of acquisition and using historical data for standard Q&A, the accuracy and efficiency of work order data processing are effectively improved, and the problem of insufficient accuracy brought about by model illusion is reduced.
Smart Images
Figure CN119621905A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a work order data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art
[0002] In the fields of customer service, technical support, and government public service platforms, work order systems are widely used to record problems, requests, and solutions submitted by customers or the public. In related technologies, the intelligent processing of work order data mainly relies on two methods: fine-tuning based on large language models and retrieval-based question-answering systems.
[0003] However, the fine-tuning based on large language models involved in the related art has the problem of difficulty in updating knowledge, and the obtained model may generate questions and answers that are inconsistent with the facts. The retrieval-based question-answering system involved in the related art provides answers by retrieving old work orders related to new work orders. This method can provide answers, but the knowledge is fragmented, the response time is long, and there is also the risk of model hallucination, and the retrieval results are not accurate enough. Summary of the invention
[0004] Based on this, it is necessary to provide a work order data processing method, device, computer equipment, computer readable storage medium and computer program product that can improve data processing accuracy in response to the above technical problems.
[0005] In a first aspect, the present application provides a work order data processing method, comprising:
[0006] Obtain the work order data to be processed, and perform question-answer pair extraction on the work order data to be processed, to obtain the question-answer pairs to be processed corresponding to the work order data to be processed;
[0007] Determine a first standard question-answer pair according to the question-answer pair to be processed and the historical question-answer pairs;
[0008] Determine the second standard question-answer pair based on the pending work order data and the historical work order data;
[0009] According to the first standard question-answer pair and the second standard question-answer pair, a target standard question-answer pair corresponding to the work order data to be processed is determined.
[0010] In one embodiment, a first standard question and answer pair is determined based on a first question and answer pair to be processed and a historical question and answer pair, including: calculating a first similarity between the first question and answer pair to be processed and the historical question and answer pair, and selecting a historical question and answer pair whose first similarity meets a first requirement as an initial question and answer pair; and determining a first standard question and answer pair based on the initial question and answer pair and a pre-generated question and answer pair relationship, wherein the question and answer pair relationship is used to characterize the correspondence between the historical question and answer pairs and the standard question and answer pairs.
[0011] In one of the embodiments, a second standard question-answer pair is determined based on the work order data to be processed and the historical work order data, including: calculating a second similarity between the work order data to be processed and the historical work order data, and selecting the historical work order data whose second similarity meets the second requirement as the initial work order data; and determining a second standard question-answer pair based on a pre-generated work order question-answer pair relationship and the initial work order data, wherein the work order question-answer pair relationship is used to characterize the correspondence between the historical work order data and the standard question-answer pair.
[0012] In an optional embodiment, a target standard question and answer pair corresponding to the work order data to be processed is determined based on the first standard question and answer pair and the second standard question and answer pair, including: determining a weight value of the first standard question and answer pair corresponding to the initial question and answer pair based on the first similarity; determining a weight value of the second standard question and answer pair corresponding to the initial work order data based on the second similarity; and determining the target standard question and answer pair based on the weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair.
[0013] In one of the embodiments, after determining the weight value of the second standard question and answer pair corresponding to the initial work order data based on the second similarity, the method also includes: obtaining attribute information of the work order data to be processed; when the attribute information of the work order data to be processed matches the attribute information of the initial work order data, increasing the weight value corresponding to the second standard question and answer pair; determining the target standard question and answer pair based on the weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair, including: determining the target standard question and answer pair based on the weight value of the first standard question and answer pair and the increased weight value of the second standard question and answer pair.
[0014] In one embodiment, after determining the target standard question and answer pair corresponding to the work order data to be processed, the method also includes: obtaining feedback information for the target standard question and answer pair, the feedback information is used to indicate whether the target standard question and answer pair matches the work order data to be processed; when the feedback information indicates that the target standard question and answer pair matches the work order data to be processed, using the target standard question and answer pair and the work order data to be processed to update historical data, the historical data including historical work order data and historical question and answer pairs.
[0015] In a second aspect, the present application also provides a work order data processing device, comprising:
[0016] The work order data acquisition module is used to obtain the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed;
[0017] A first determination module, configured to determine a first standard question-answer pair based on the question-answer pair to be processed and the historical question-answer pairs;
[0018] A second determination module, used to determine a second standard question-answer pair based on the to-be-processed work order data and the historical work order data;
[0019] The target determination module is used to determine the target standard question and answer pair corresponding to the work order data to be processed based on the first standard question and answer pair and the second standard question and answer pair.
[0020] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.
[0021] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned method embodiments when the computer program is executed by a processor.
[0022] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned method embodiments when executed by a processor.
[0023] The above-mentioned work order data processing method, device, computer equipment, computer-readable storage medium and computer program product determine the first standard question and answer pair based on the question and answer pair to be processed and the historical question and answer pair, and determine the second standard question and answer pair based on the work order data to be processed and the historical work order data, and implement the processing of the target standard question and answer pair according to the first standard question and answer pair and the second standard question and answer pair. In the process of obtaining the target standard question and answer pair, not only the question and answer pair is used to obtain the first standard question and answer pair, but also the work order data is used to obtain the second standard question and answer pair, and then the first standard question and answer pair and the second standard question and answer pair are combined to achieve the acquisition of the target standard question and answer pair. On the one hand, the scope of obtaining the standard question and answer pair is expanded, and the scope of the target standard question and answer pair is finally obtained, thereby increasing the possibility of obtaining accurate target standard question and answer pairs; on the other hand, in the process of obtaining the standard question and answer pair, historical data is used for processing, and processing is performed based on the similarity between the historical data and the work order data to be processed, so as to reduce the problem of insufficient accuracy due to model hallucination and effectively improve the accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0025] Figure 1 An application environment diagram of a work order data processing method in one embodiment;
[0026] Figure 2 A schematic diagram of a flow chart of a work order data processing method in one embodiment;
[0027] Figure 3 A schematic diagram of a flow chart of a work order data processing method in another embodiment;
[0028] Figure 4 is a structural block diagram of a work order data processing device in one embodiment;
[0029] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0030] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0031] The work order data processing method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 sends the work order data to be processed to the server 104, and the server 104 obtains the work order data to be processed, and extracts the question and answer pairs of the work order data to be processed to obtain the question and answer pairs to be processed corresponding to the work order data to be processed; according to the question and answer pairs to be processed and the historical question and answer pairs, the first standard question and answer pairs are determined; according to the work order data to be processed and the historical work order data, the second standard question and answer pairs are determined; according to the first standard question and answer pairs and the second standard question and answer pairs, the target standard question and answer pairs corresponding to the work order data to be processed are determined. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.
[0032] In an exemplary embodiment, Figure 2 As shown, a work order data processing method is provided, and the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 202 to 206. Among them:
[0033] Step 202, obtain the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed.
[0034] Among them, work order data is used to represent work documents.
[0035] Optionally, the work order data can be used to record questions, requests and solutions submitted by customers or the public, and operators can process based on the content recorded in the work order data, for example, answering questions recorded in the work order data.
[0036] It is understandable that the work order data may include noise data that is not related to the problem or request. In this case, the work order data can be processed by question-answer pair extraction so as to process the work order data in a targeted manner and improve the efficiency and accuracy of data processing.
[0037] Optionally, the question-answer pair to be processed may include the question to be processed and the answer corresponding to the question to be processed.
[0038] In one embodiment, a large language model can be combined with a prompt project to extract question-answer pairs from the work order data to be processed, so as to obtain question-answer pairs to be processed. Optionally, the large language model can be GPT, Qwen, etc.
[0039] Among them, Prompt Engineering refers to guiding the model to directly adapt to and perform specific downstream tasks through carefully designed prompts based on pre-trained large language models (such as GPT, Qwen, etc.).
[0040] Optionally, a large language model may be used in combination with a question-answer pair extraction prompt project to implement question-answer pair extraction processing on the work order data to be processed.
[0041] In one embodiment, when the work order data to be processed is question or request related data, a large language model is used in combination with a question-answer pair extraction prompt project to implement question-answer pair extraction processing on the work order data to be processed. The obtained question-answer pairs to be processed may only include questions, and the corresponding answers are zero.
[0042] Optionally, when the work order data to be processed includes questions, requests and their corresponding solutions, a large language model is used in combination with a question-answer pair extraction prompt project to implement question-answer pair extraction processing on the work order data to be processed. The obtained question-answer pairs to be processed may include both questions and corresponding answers.
[0043] In one embodiment, the question-answer pair to be processed obtained by performing question-answer pair extraction processing on the work order data to be processed may be one or more question-answer pairs to be processed.
[0044] Step 204: Determine a first standard question-answer pair based on the question-answer pair to be processed and the historical question-answer pairs.
[0045] In one embodiment, the first standard question-answer pair may be determined based on the similarity between the question-answer pair to be processed and the historical question-answer pairs.
[0046] Optionally, a first similarity between the first question-and-answer pair to be processed and the historical question-and-answer pair is calculated, and a historical question-and-answer pair whose first similarity meets a first requirement is selected as an initial question-and-answer pair; and a first standard question-and-answer pair is determined based on the initial question-and-answer pair and the pre-generated question-and-answer pair relationship.
[0047] Among them, the question-answer pair relationship is used to represent the correspondence between historical question-answer pairs and standard question-answer pairs.
[0048] Optionally, you can obtain the unprocessed question included in the first question-and-answer pair to be processed, calculate the similarity between the unprocessed question included in the first question-and-answer pair to be processed and the historical question included in the historical question-and-answer pair, determine the similarity value between the unprocessed question and the historical question included in the historical question-and-answer pair, and determine the initial question-and-answer pair based on the similarity value.
[0049] In one embodiment, the first requirement may be a first preset number. Optionally, the first preset number is obtained, and the initial question-answer pairs are obtained according to the first preset number. The first preset number is used to represent the number of question-answer pairs obtained.
[0050] Optionally, the initial question-answer pair is determined according to a first preset number and a similarity value between the question to be processed and the historical question included in the historical question-answer pair. Exemplarily, the historical question-answer pairs are sorted according to the similarity value, a first preset number of historical question-answer pairs are obtained, and the first preset number of historical question-answer pairs obtained are used as the initial question-answer pairs.
[0051] In one embodiment, the first requirement may also be a first preset similarity threshold. Optionally, the first preset similarity threshold may be set, and the initial question-answer pair may be determined based on the first preset similarity threshold and the first similarity between the question to be processed and the historical question included in the historical question-answer pair. Exemplarily, a historical question-answer pair having a first similarity greater than the first preset similarity threshold is obtained, and the obtained historical question-answer pair is used as the initial question-answer pair.
[0052] In an optional embodiment, after the initial question-answer pair is obtained, the question-answer pair relationship may be queried to determine the first standard question-answer pair corresponding to the initial question-answer pair.
[0053] In one of the embodiments, question-answer pair relationships may be pre-generated so that standard question-answer pairs may be directly acquired through the question-answer pair relationships when needed.
[0054] Optionally, historical data is obtained and standard question-answer pairs are generated based on the historical data.
[0055] In one embodiment, before generating standard question-answer pairs based on historical data, the historical data may be preprocessed to reduce historical data noise and improve data processing efficiency. For example, advertisements, templated greetings, etc. in the historical data may be removed.
[0056] In one embodiment, when preprocessing the historical data, the historical data may also be desensitized to protect personal privacy and sensitive information. For example, personal information such as name, phone number, address, etc. included in the historical data is replaced.
[0057] In one embodiment, if the historical data includes multimedia format data such as pictures, audio, and video, the preprocessing of the historical data may also include assisting in describing the historical data with the aid of technologies such as computer vision and speech recognition. Optionally, assisting in describing the historical data with the aid of technologies such as computer vision and speech recognition may generate a text file corresponding to the historical data.
[0058] In one embodiment, after preprocessing the historical data, historical work order data can be obtained, and generation of question-answer pair relationships is achieved based on the historical work order data.
[0059] Optionally, a large language model can be used in combination with a question-answer pair extraction prompt project to implement question-answer pair extraction processing on historical work order data to obtain historical question-answer pairs corresponding to the historical work order data. Optionally, the large language model can be GPT, Qwen, etc.
[0060] In one embodiment, the historical question-answer pairs corresponding to the historical work order data may be one or more.
[0061] In an optional embodiment, the generation of standard question-answer pairs may be achieved based on historical question-answer pairs corresponding to historical work order data.
[0062] It is understandable that in order to ensure the accuracy of the generation of standard question and answer pairs, before using historical question and answer pairs to generate standard question and answer pairs, quality inspection can be performed on historical question and answer pairs corresponding to historical work orders to determine the accuracy of the historical question and answer pairs.
[0063] Optionally, a large language model can be used in combination with a quality detection prompt project to implement quality detection of question-answer pairs on historical work order data to determine whether the historical question-answer pairs are accurate. Optionally, the large language model can be GPT, Qwen, etc. Optionally, detecting the accuracy of historical question-answer pairs can include detecting whether the historical questions included in the historical question-answer pairs are accurately stated, and whether the historical answers corresponding to the historical questions are related to the historical questions and are accurate.
[0064] In one embodiment, when the quality detection is inaccurate, a large language model can be used in combination with an extended rewriting prompt project to rewrite and expand historical question and answer pairs to obtain accurate historical question and answer pairs, and the historical questions and answers obtained after rewriting and expansion are set corresponding to historical work orders to obtain the relationship between historical work orders and historical question and answer pairs.
[0065] Optionally, when the quality inspection is accurate, the historical work order pairs and the historical work orders are set correspondingly to obtain the relationship between the historical work orders and the historical question and answer pairs.
[0066] In one embodiment, after obtaining the relationship between historical work orders and historical question-answer pairs, a topic model (such as an LDA model) may be used to process the historical work orders to determine the topic corresponding to each historical work order.
[0067] Optionally, clustering processing can be performed based on the topics corresponding to each historical work order to form a historical work order cluster, and each historical work order data in the same historical work order cluster can be marked, so that each historical work order data in the same historical work order cluster, and the historical question and answer pairs corresponding to each historical work order data have a corresponding relationship. The historical work order data can be used to query other historical work order data and historical question and answer pairs corresponding to other historical work order data. Similarly, other historical work order data and historical question and answer pairs corresponding to other historical work order data can be queried.
[0068] In one embodiment, the topic model can also be used to implement topic acquisition processing for the historical questions included in each historical question-answer pair, and cluster processing is performed based on the topics corresponding to each acquired historical question to obtain a historical question cluster. Optionally, each historical question in the obtained historical question cluster is marked so that any historical question can be used to obtain other historical questions, and historical work orders and historical answers can be obtained based on other acquired historical questions.
[0069] In an optional embodiment, after obtaining the historical work order clusters and / or historical question and answer clusters, the historical work order clusters and / or historical question and answer clusters can be output, and the historical work order clusters and / or historical question and answer clusters can be verified and corrected by an operator.
[0070] Optionally, the operator may also generate standard question and answer pairs for historical work order clusters and / or historical question and answer pair clusters during the verification and correction process to form question and answer pair relationships and / or work order question and answer pair relationships.
[0071] Exemplarily, the historical question-and-answer pairs corresponding to the historical work orders at the cluster center in the historical work order cluster can be used as standard question-and-answer pairs, or question-and-answer pairs based on the topics corresponding to the historical work order cluster can be input as standard question-and-answer pairs, and the standard question-and-answer pairs can be set corresponding to the historical work order clusters to form a work order question-and-answer pair relationship, which is used to characterize the correspondence between historical work order data and standard question-and-answer pairs.
[0072] Exemplarily, the historical question and answer pairs at the cluster center in the historical question and answer pair cluster can be used as standard question and answer pairs, or question and answer pairs based on the topics corresponding to the historical question and answer pair cluster can be input as standard question and answer pairs, and the standard question and answer pairs can be set corresponding to the historical question and answer pair cluster to form a question and answer pair relationship.
[0073] Step 206: Determine a second standard question-answer pair based on the work order data to be processed and the historical work order data.
[0074] In one embodiment, the second standard question-answer pair may be determined by utilizing the similarity between the work order data to be processed and the historical work order data.
[0075] Optionally, a second similarity between the work order data to be processed and the historical work order data is calculated, and the historical work order data whose second similarity meets the second requirement is selected as the initial work order data; and a second standard question and answer pair is determined based on the pre-generated work order question and answer pair relationship and the initial work order data.
[0076] Optionally, the topic model can be used to process the work order data to obtain the pending topics corresponding to the work order data to be processed, and similarity calculation can be performed based on the pending topics and the historical topics corresponding to the historical work order data to determine the second similarity between the work order data to be processed and the historical work order data, and the initial work order data can be determined based on the second similarity value.
[0077] In one embodiment, the second requirement may be a second preset number. Optionally, the second preset number is obtained, and the initial work order is obtained according to the second preset number. The second preset number is used to represent the number of initial work orders obtained.
[0078] Optionally, the initial work order data is determined according to a second preset number and a second similarity value. Exemplarily, the historical work order data is sorted according to the second similarity, a second preset number of historical work order data is obtained, and the obtained second preset number of historical work order data is used as the initial work order data.
[0079] In one embodiment, the second requirement may also be a second preset similarity threshold. Optionally, a second preset similarity threshold may be set, and the initial work order data may be determined based on the second similarity and the second similarity. Exemplarily, historical work orders having a second similarity to the second preset similarity threshold are obtained, and the obtained historical work order data are used as the initial work order data.
[0080] In an optional embodiment, after the initial work order data is acquired, the work order question-answer pair relationship may be queried to determine the second standard question-answer pair corresponding to the initial work order data.
[0081] In one embodiment, the first preset number and the second preset number may be the same or different.
[0082] In one embodiment, the first preset similarity threshold and the second similarity threshold may be the same or different.
[0083] Step 208: Determine a target standard question-answer pair corresponding to the work order data to be processed based on the first standard question-answer pair and the second standard question-answer pair.
[0084] In one embodiment, based on the first similarity, the weight value of the first standard question and answer pair corresponding to the initial question and answer pair is determined; based on the second similarity, the weight value of the second standard question and answer pair corresponding to the initial work order data is determined; based on the weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair, the target standard question and answer pair is determined.
[0085] Optionally, the correspondence between similarity and weight value can be pre-generated, so that after the similarity is determined, the weight value corresponding to the first standard question and answer pair and the weight value corresponding to the second standard question and answer pair can be determined based on the correspondence between the similarity and the weight value, and the target standard question and answer pair can be determined based on the determined weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair.
[0086] In one embodiment, the first standard question-answer pair and the second standard question-answer pair may also be ranked by similarity. Optionally, a correspondence between the ranking and the weight value is pre-generated, and after the ranking of the first standard question-answer pair and the second standard question-answer pair is determined, the weight values of the first standard question-answer pair and the second standard question-answer pair are determined according to the correspondence between the ranking and the weight value.
[0087] In one embodiment, the corresponding relationship between the similarity and the weight value corresponding to the question and answer pair and the corresponding relationship between the similarity and the weight value corresponding to the historical work order data may be different or the same.
[0088] In one embodiment, the corresponding relationship between the ranking and weight value corresponding to the question and answer pairs and the corresponding relationship between the ranking and weight value corresponding to the historical work order data may be different or the same.
[0089] In an optional embodiment, after determining the weight value of the second standard question and answer pair corresponding to the initial work order data based on the second similarity, the method also includes: obtaining attribute information of the work order data to be processed; when the attribute information of the work order data to be processed matches the attribute information of the initial work order data, increasing the weight value corresponding to the second standard question and answer pair; determining the target standard question and answer pair based on the weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair, including: determining the target standard question and answer pair based on the weight value of the first standard question and answer pair and the increased weight value of the second standard question and answer pair.
[0090] The attribute information is used to characterize the characteristic information of the work order data to be processed. Optionally, the attribute information may include but is not limited to user information, geographic location, and time information.
[0091] Optionally, the attribute information of the work order data to be processed and the attribute information of the initial work order data are obtained, and the attribute information of the work order data to be processed and the attribute information of the initial work order data are processed for similarity to determine whether the attribute information of the work order data to be processed is similar to the attribute information of the initial work order data, and the weight value can be adjusted according to the similarity.
[0092] Exemplarily, if the attribute information of the work order data to be processed is similar to the attribute information of the initial work order data, the weight value of the initial work order data may be increased.
[0093] Optionally, if the attribute information of the work order data to be processed is not similar to the attribute information of the initial work order data, the weight value of the initial work order data may be adjusted.
[0094] Optionally, if the attribute information of the work order data to be processed is not similar to the attribute information of the initial work order data, the weight value of the initial work order data may also be adjusted to be reduced.
[0095] Optionally, a preset similarity threshold may be used. If the similarity between the attribute information of the work order data to be processed and the attribute information of the initial work order data is less than the preset similarity threshold, a weight value of the initial work order data less than the similarity threshold is adjusted to decrease. If the similarity between the attribute information of the work order data to be processed and the attribute information of the initial work order data is not less than the preset similarity threshold, the weight value of the initial work order data is not adjusted.
[0096] In an optional embodiment, after the target standard question-answer pairs are determined, a preset number of target standard question-answer pairs may be selected for output.
[0097] Optionally, the historical work order corresponding to the target standard question and answer pair may be determined based on the relationship between the work order question and answer pairs, and the determined historical work order may be output as the target work order to assist the user in understanding the standard question and answer pair.
[0098] In an optional embodiment, feedback information for the target standard question and answer pair is obtained; when the feedback information indicates that the target standard question and answer pair matches the work order data to be processed, the target standard question and answer pair and the work order data to be processed are used to update historical data, and the historical data includes historical work order data and historical question and answer pairs.
[0099] Among them, the feedback information is used to indicate whether the target standard question and answer pair matches the work order data to be processed.
[0100] Optionally, a feedback mechanism may be provided, and users may utilize the feedback mechanism to realize the matching degree of the work order data to be processed and the corresponding target standard question and answer pairs.
[0101] Exemplarily, a feedback control may be formed on the terminal, and the user may use the feedback control to provide feedback on the matching status between the work order to be processed and the corresponding target standard question and answer pair.
[0102] Optionally, if the feedback information indicates that the to-be-processed work order data and the target standard question-answer pair are set correspondingly, the historical data is updated with the correspondingly set to-be-processed work order data and the target standard question-answer pair.
[0103] In the above-mentioned work order data processing method, a first standard question and answer pair is determined based on the question and answer pair to be processed and the historical question and answer pair, and a second standard question and answer pair is determined based on the work order data to be processed and the historical work order data, and the processing of the target standard question and answer pair is realized based on the first standard question and answer pair and the second standard question and answer pair. In the process of realizing the acquisition of the target standard question and answer pair, not only the question and answer pair is used to obtain the first standard question and answer pair, but also the work order data is used to obtain the second standard question and answer pair, and then the first standard question and answer pair and the second standard question and answer pair are combined to realize the acquisition of the target standard question and answer pair. On the one hand, the scope of obtaining the standard question and answer pair is expanded, and the scope of the target standard question and answer pair is finally obtained, thereby increasing the possibility of obtaining accurate target standard question and answer pairs; on the other hand, in the process of realizing the acquisition of the standard question and answer pair, historical data is used for processing, and processing is performed based on the similarity between the historical data and the work order data to be processed, so as to reduce the problem of insufficient accuracy due to model hallucination and effectively improve the accuracy.
[0104] In an exemplary embodiment, Figure 3 As shown, a work order data processing method is provided, and the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 302 to 324. Among them:
[0105] Step 302, obtain the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed.
[0106] Step 304, calculate the first similarity between the first question-and-answer pair to be processed and the historical question-and-answer pair, and select the historical question-and-answer pair whose first similarity meets the first requirement as the initial question-and-answer pair.
[0107] Step 306: Determine a first standard question-answer pair based on the initial question-answer pair and the pre-generated question-answer pair relationship.
[0108] Among them, the question-answer pair relationship is used to represent the correspondence between historical question-answer pairs and standard question-answer pairs.
[0109] Step 308 , calculating the second similarity between the work order data to be processed and the historical work order data, and selecting the historical work order data whose second similarity meets the second requirement as the initial work order data.
[0110] Step 310, determining a second standard question-answer pair based on the pre-generated work order question-answer pair relationship and the initial work order data.
[0111] Among them, the work order question-answer pair relationship is used to characterize the correspondence between historical work order data and standard question-answer pairs.
[0112] Step 312: Determine a weight value of a first standard question-answer pair corresponding to the initial question-answer pair based on the first similarity.
[0113] Step 314: Determine the weight value of the second standard question-answer pair corresponding to the initial work order data according to the second similarity.
[0114] Step 316, obtaining attribute information of the work order data to be processed.
[0115] Step 318: When the attribute information of the work order data to be processed matches the attribute information of the initial work order data, the weight value corresponding to the second standard question and answer pair is increased.
[0116] Step 320: Determine a target standard question-answer pair based on the weight value of the first standard question-answer pair and the weight value of the second standard question-answer pair.
[0117] Step 322, obtaining feedback information for the target standard question and answer pair.
[0118] Among them, the feedback information is used to indicate whether the target standard question and answer pair matches the work order data to be processed.
[0119] Step 324 , determine whether the feedback information indicates that the target standard question and answer pair matches the work order data to be processed. If so, execute step 326 ; if not, execute step 328 .
[0120] Step 326, update the historical data using the target standard question and answer pairs and the work order data to be processed.
[0121] Among them, historical data includes historical work order data and historical question and answer pairs.
[0122] Step 328, receiving the adjusted target standard question and answer pair for the work order data to be processed, and using the adjusted target standard question and answer pair and the corresponding work order data to be processed to update the historical data.
[0123] In this embodiment, a first standard question and answer pair is determined based on the question and answer pair to be processed and the historical question and answer pairs, and a second standard question and answer pair is determined based on the work order data to be processed and the historical work order data, and the processing of the target standard question and answer pair is realized based on the first standard question and answer pair and the second standard question and answer pair. In the process of realizing the acquisition of the target standard question and answer pair, not only the question and answer pair is used to obtain the first standard question and answer pair, but also the work order data is used to obtain the second standard question and answer pair, and then the first standard question and answer pair and the second standard question and answer pair are combined to realize the acquisition of the target standard question and answer pair. On the one hand, the scope of obtaining the standard question and answer pair is expanded, and the scope of the target standard question and answer pair is finally obtained, thereby improving the possibility of obtaining accurate target standard question and answer pairs; on the other hand, in the process of realizing the acquisition of the standard question and answer pair, historical data is used for processing, and the processing is based on the similarity between the historical data and the work order data to be processed, so as to reduce the problem of insufficient accuracy due to model hallucination and effectively improve the accuracy.
[0124] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0125] Based on the same inventive concept, the embodiment of the present application also provides a work order data processing device for implementing the work order data processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more work order data processing device embodiments provided below can refer to the limitations on the work order data processing method above, and will not be repeated here.
[0126] In an exemplary embodiment, Figure 4 As shown, a work order data processing device 400 is provided, comprising: a work order data acquisition module 402, a first determination module 404, a second determination module 406 and a target determination module 408, wherein:
[0127] The work order data acquisition module 402 is used to acquire the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed.
[0128] The first determination module 404 is used to determine a first standard question-answer pair according to the question-answer pair to be processed and the historical question-answer pairs.
[0129] The second determination module 406 is used to determine a second standard question-answer pair according to the to-be-processed work order data and the historical work order data.
[0130] The target determination module 408 is used to determine the target standard question and answer pair corresponding to the work order data to be processed according to the first standard question and answer pair and the second standard question and answer pair.
[0131] In one embodiment, the first determination module is also used to calculate the first similarity between the first question and answer pair to be processed and the historical question and answer pair, and select the historical question and answer pair whose first similarity meets the first requirement as the initial question and answer pair; determine the first standard question and answer pair based on the initial question and answer pair and the pre-generated question and answer pair relationship, wherein the question and answer pair relationship is used to characterize the correspondence between the historical question and answer pair and the standard question and answer pair.
[0132] In one of the embodiments, the second determination module is also used to calculate the second similarity between the work order data to be processed and the historical work order data, and select the historical work order data whose second similarity meets the second requirement as the initial work order data; determine the second standard question and answer pair based on the pre-generated work order question and answer pair relationship and the initial work order data, wherein the work order question and answer pair relationship is used to characterize the correspondence between the historical work order data and the standard question and answer pair.
[0133] In an optional embodiment, the target determination module is used to determine the weight value of the first standard question and answer pair corresponding to the initial question and answer pair based on the first similarity; determine the weight value of the second standard question and answer pair corresponding to the initial work order data based on the second similarity; and determine the target standard question and answer pair based on the weight value of the first standard question and answer pair and the weight value of the second standard question and answer pair.
[0134] In one embodiment, the work order data processing device also includes a weight adjustment module for obtaining attribute information of the work order data to be processed; when the attribute information of the work order data to be processed matches the attribute information of the initial work order data, the weight value corresponding to the second standard question and answer pair is increased.
[0135] The target determination module is also used to determine the target standard question and answer pair based on the weight value of the first standard question and answer pair and the increased weight value of the second standard question and answer pair.
[0136] In one embodiment, the work order data processing device also includes a feedback module, which is also used to obtain feedback information for the target standard question and answer pair, and the feedback information is used to indicate whether the target standard question and answer pair matches the work order data to be processed; when the feedback information indicates that the target standard question and answer pair matches the work order data to be processed, the target standard question and answer pair and the work order data to be processed are used to update the historical data, and the historical data includes historical work order data and historical question and answer pairs.
[0137] Each module in the work order data determination device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.
[0138] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a work order data processing method is implemented.
[0139] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0140] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.
[0141] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method embodiments are implemented.
[0142] In one embodiment, a computer program product is provided, including a computer program, which implements the steps of the above method embodiments when executed by a processor.
[0143] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0144] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.
[0145] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0146] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A work order data processing method, characterized in that: The method comprises: Acquire the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed; Determine a first standard question-answer pair according to the to-be-processed question-answer pair and the historical question-answer pairs; Determining a second standard question-answer pair according to the to-be-processed work order data and the historical work order data; According to the first standard question and answer pair and the second standard question and answer pair, a target standard question and answer pair corresponding to the work order data to be processed is determined.
2. The method according to claim 1, characterized in that Determining a first standard question-answer pair according to the first to-be-processed question-answer pair and the historical question-answer pair includes: Calculating a first similarity between the first to-be-processed question-answer pair and the historical question-answer pair, and selecting a historical question-answer pair whose first similarity meets a first requirement as an initial question-answer pair; The first standard question-answer pair is determined based on the initial question-answer pair and the pre-generated question-answer pair relationship, wherein the question-answer pair relationship is used to characterize the correspondence between the historical question-answer pair and the standard question-answer pair.
3. The method according to claim 2, characterized in that Determining a second standard question-answer pair according to the to-be-processed work order data and the historical work order data includes: Calculating a second similarity between the to-be-processed work order data and the historical work order data, and selecting the historical work order data whose second similarity meets a second requirement as initial work order data; The second standard question and answer pair is determined according to the pre-generated work order question and answer pair relationship and the initial work order data, wherein the work order question and answer pair relationship is used to characterize the correspondence between the historical work order data and the standard question and answer pair.
4. The method according to claim 3, characterized in that The step of determining a target standard question-answer pair corresponding to the to-be-processed work order data according to the first standard question-answer pair and the second standard question-answer pair includes: Determining a weight value of a first standard question-answer pair corresponding to the initial question-answer pair according to the first similarity; Determining a weight value of a second standard question-answer pair corresponding to the initial work order data according to the second similarity; The target standard question-answer pair is determined according to the weight value of the first standard question-answer pair and the weight value of the second standard question-answer pair.
5. The method according to claim 4, characterized in that After determining the weight value of the second standard question-answer pair corresponding to the initial work order data according to the second similarity, the method further includes: Obtaining attribute information of the work order data to be processed; When the attribute information of the to-be-processed work order data matches the attribute information of the initial work order data, increasing the weight value corresponding to the second standard question-answer pair; The step of determining the target standard question-answer pair according to the weight value of the first standard question-answer pair and the weight value of the second standard question-answer pair includes: The target standard question and answer pair is determined according to the weight value of the first standard question and answer pair and the increased weight value of the second standard question and answer pair.
6. The method according to any one of claims 1 to 5, characterized in that: After determining the target standard question-answer pair corresponding to the work order data to be processed, the method further includes: Acquire feedback information for the target standard question-answer pair, where the feedback information is used to indicate whether the target standard question-answer pair matches the work order data to be processed; When the feedback information indicates that the target standard question and answer pair matches the work order data to be processed, the historical data is updated using the target standard question and answer pair and the work order data to be processed, and the historical data includes historical work order data and historical question and answer pairs.
7. A work order data processing device, characterized in that: The device comprises: A work order data acquisition module is used to acquire the work order data to be processed, and perform question-answer pair extraction processing on the work order data to be processed to obtain the question-answer pairs to be processed corresponding to the work order data to be processed; A first determination module, configured to determine a first standard question-answer pair based on the question-answer pair to be processed and the historical question-answer pairs; A second determination module, configured to determine a second standard question-answer pair based on the to-be-processed work order data and the historical work order data; A target determination module is used to determine a target standard question and answer pair corresponding to the work order data to be processed based on the first standard question and answer pair and the second standard question and answer pair.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.