Feedback processing method and device, electronic equipment and storage medium
Through multi-level tag chain matching and automatic distribution technology, the problem of rapid positioning and low processing efficiency of feedback problems in user feedback processing is solved, and the rapid distribution and processing of feedback information is realized, improving user experience and processing efficiency.
Patent Information
- Application Number
- CN202411017045.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-26
AI Technical Summary
During the user feedback processing process, different personnel are responsible for the maintenance of faults in different links, which leads to the rapid positioning and low processing efficiency of feedback problems, affecting product optimization and brand image.
By obtaining target feedback information, determining its corresponding multi-level tag chain, and matching it with the tag chain in the historical feedback database, the feedback information is automatically sent to the corresponding client, thereby achieving the rapid distribution of feedback information and the rapid positioning of processors .
It reduces the time-consuming and efficient feedback processing, optimizes user experience, and improves the accuracy and efficiency of feedback processing.
Smart Images

Figure CN119003796A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of computer technology, in particular to the field of artificial intelligence technology such as natural language processing, large language models, and deep learning, and specifically to a feedback processing method, device, electronic device, and storage medium. Background Art
[0002] When an application is used by users, customer complaints are usually generated due to application failures at different links. Timely handling of customer complaints and maintenance of application failures are the key to promoting the continuous optimization and upgrading of products and services and maintaining the brand image of the company. Since maintenance of failures at different links is usually handled by different personnel, how to quickly locate the person responsible for user feedback issues is the key to improving the efficiency of feedback processing. Summary of the invention
[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.
[0004] To this end, the purpose of the present disclosure is to propose a feedback processing method, device, electronic device and storage medium to realize the automatic transfer of user feedback issues to specific processing personnel, reduce the time consumption of the feedback flow link, and improve the efficiency of feedback processing.
[0005] According to a first aspect of the present disclosure, a feedback processing method is provided, comprising:
[0006] Obtain target feedback information to be processed;
[0007] Determine a first label chain corresponding to the target feedback information;
[0008] Matching the first label chain with each historical label chain in a historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain;
[0009] The target feedback information is sent to the first client.
[0010] According to a second aspect of the present disclosure, a feedback processing device is provided, comprising:
[0011] An acquisition module, used to acquire target feedback information to be processed;
[0012] A first determining module, used to determine a first label chain corresponding to the target feedback information;
[0013] A second determination module is used to match the first label chain with each historical label chain in a historical feedback database respectively, to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain;
[0014] A sending module is used to send the target feedback information to the first client.
[0015] According to a third aspect of the present disclosure, there is provided an electronic device, including:
[0016] at least one processor; and
[0017] a memory communicatively connected to the at least one processor; wherein,
[0018] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the feedback processing method as described in the first aspect.
[0019] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the feedback processing method as described in the first aspect.
[0020] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising computer instructions, which, when executed by a processor, implement the steps of the feedback processing method as described in the first aspect.
[0021] The feedback processing method, device, electronic device and storage medium provided by the present disclosure have the following beneficial effects:
[0022] By multi-level classification of the target feedback information, the first label chain is obtained, and then the client identifier corresponding to the historical label chain matching the target feedback information is obtained based on the first label chain, so that the target feedback information is sent to the client corresponding to the client identifier to complete the distribution of the feedback information. The rapid positioning and automatic distribution of the user feedback processing personnel are realized, the time consumption of the feedback circulation link is reduced, the efficiency of feedback processing is improved, and the user experience is optimized.
[0023] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, which are used to better understand the present solution and do not constitute a limitation of the present disclosure, wherein:
[0025] Figure 1 is a flowchart of a feedback processing method proposed according to an embodiment of the present disclosure;
[0026] Figure 2 is a flowchart of a feedback processing method proposed according to another embodiment of the present disclosure;
[0027] Figure 3 is a flowchart of a feedback processing method proposed according to another embodiment of the present disclosure;
[0028] Figure 4 is a flowchart of a feedback processing method proposed according to another embodiment of the present disclosure;
[0029] Figure 5 is a structural diagram of a feedback processing device proposed according to an embodiment of the present disclosure;
[0030] Figure 6 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION
[0031] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.
[0032] The embodiments of the present disclosure relate to artificial intelligence technology fields such as natural language processing, large language models, and deep learning.
[0033] Natural Language Understanding (NLU) is a general term for all methods, models or tasks that support machines to understand text content. NLU plays a very important role in text information processing systems and is an essential module for recommendation, question-answering, search and other systems.
[0034] Large Language Model (LLM) refers to a deep learning model trained with a large amount of text data, which can generate natural language text or understand the meaning of language text. Large language models can handle a variety of natural language tasks, such as text classification, question answering, and dialogue, and are an important path to artificial intelligence.
[0035] Deep learning is the process of learning the inherent laws and representation levels of sample data. The information obtained in the learning process is of great help in interpreting data such as text, images, and sounds. Its ultimate goal is to enable machines to have analytical learning capabilities like humans and to be able to recognize data such as text, images, and sounds.
[0036] Artificial Intelligence, abbreviated as AI in English, is an important part of the intelligent discipline. It attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is a very broad science, including robots, language recognition, image recognition, natural language processing, expert systems, machine learning, computer vision, etc.
[0037] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0038] The feedback processing method, device, electronic device, and storage medium of the embodiments of the present disclosure are described below with reference to the accompanying drawings.
[0039] It should be noted that the executor of the feedback processing method of this embodiment is a feedback processing device, which can be implemented by software and / or hardware. The device can be configured in an electronic device, and the electronic device can include but is not limited to a terminal, a server, etc.
[0040] Figure 1 It is a flowchart of a feedback processing method proposed according to an embodiment of the present disclosure.
[0041] like Figure 1 As shown, the feedback processing method includes:
[0042] S101: Obtain target feedback information to be processed.
[0043] Among them, the target feedback information may include feedback content used to describe the specific problem in detail, and a feedback title that summarizes or concludes the feedback content. The feedback title may be a title summarized and entered by the customer service after the customer receives the feedback content entered by the user in the dialogue window.
[0044] In the embodiment of the present application, when the customer service stores a feedback information in the user feedback pool, it can determine that the target feedback information to be processed is obtained and directly process it. Alternatively, a thread can be established in the server to query whether there is unprocessed feedback information in the user feedback pool at regular intervals, and start processing it as the target feedback information. The time interval between each two queries can be determined according to the actual application maintenance needs, for example, it can be 3 minutes.
[0045] S102: Determine a first label chain corresponding to target feedback information.
[0046] The first label chain refers to a chain structure composed of multiple levels of labels used to represent the classification of the target feedback information. The number of levels of labels in the first label chain can be determined according to the complexity of the feedback problem, the accuracy requirements of the feedback processing, etc. For example, the number of levels can be 3, 5, etc., which is not limited in this disclosure.
[0047] In addition, as the number of levels of multi-level tags increases, the scope of the tag description becomes smaller. For example, a level 1 tag can be a scene tag, such as transmission, storage, account, activity, etc.; a level 2 tag can be a function tag, such as upload, backup, decompression, etc.; a level 3 tag can be a problem tag, such as upload failure, backup failure, etc.
[0048] In the disclosed embodiment, the target feedback information can be classified level by level according to the pre-set multi-level labels to obtain each level label that matches the target feedback information, and then each level label is organized into a chain structure according to its corresponding level number, so as to determine the first label chain corresponding to the target feedback information.
[0049] Optionally, multiple levels of candidate tags and the associations between the candidate tags can be obtained first. Then, the target feedback information is matched with each level of candidate tags to determine each level of tags that matches the target feedback information. After that, the tags at each level that match the target feedback information can be spliced according to the associations between the candidate tags at each level to obtain a first tag chain.
[0050] In the disclosed embodiment, a tag database may be pre-constructed, and the tag database may include multiple levels of candidate tags, each level may include multiple candidate tags, and the previous level candidate tag may be associated with one or more next level candidate tags.
[0051] In the disclosed embodiment, the target feedback information can be semantically analyzed in a variety of ways (such as a large language model and other models or algorithms for semantic recognition) to determine which label the target feedback label matches step by step, so as to complete the matching between the target feedback information and each level of candidate labels. Then, the matched candidate labels can be combined into a chain structure according to their association relationship in the label database to obtain the first label chain.
[0052] For example, the target feedback information matches the label in the level 1 candidate label as transmission, the label in the level 2 candidate label as backup, and the label in the level 3 candidate label as backup failure. Based on the association between these three labels, the first label chain can be obtained as transmission-backup-backup failure.
[0053] It should be noted that the target feedback information can match multiple candidate labels at the same level, and there may be one or more first label chains finally obtained. In the case of multiple first label chains, the levels of labels contained in any two first label chains may also be different.
[0054] In the disclosed embodiment, by matching the target feedback information with the candidate tags, obtaining the matched candidate tags and linking them in the order of association of the candidate tags to obtain the first tag chain, the accuracy and efficiency of the target feedback information classification can be improved, and a more accurate basis for determining the flow direction of the target feedback information is provided.
[0055] S103: Match the first label chain with each historical label chain in the historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain.
[0056] Among them, the historical feedback database is used to store data generated after processing historically received user feedback, which may include a historical label chain for classifying and annotating historical user feedback, as well as the content, reproduction path, problem attribution and solution of historical user feedback.
[0057] The first client identifier refers to the identifier of the client used by R&D engineers and other staff responsible for resolving user feedback.
[0058] In the disclosed embodiment, the first client identifier corresponding to each historical tag chain can be stored in the historical feedback database to record the staff responsible for processing the user feedback corresponding to the historical tag chain. It should be noted that the first client identifiers corresponding to the two historical tag chains may be the same or different.
[0059] In an embodiment of the present disclosure, the first label chain can be matched with each historical label chain in the historical feedback database step by step. When the labels in any historical label chain match the labels in the first label chain, the any historical label chain is determined as a reference label chain, so that the first client identifier corresponding to the reference label chain can be determined in the historical feedback database.
[0060] S104: Send the target feedback information to the first client.
[0061] In the disclosed embodiment, after obtaining the first client identifier corresponding to the reference label chain, since the reference label chain matches the first label chain corresponding to the target feedback information, the processing methods of the target feedback information and the historical feedback corresponding to the reference label chain should be similar. In this way, the first client identifier can be determined as the identifier of the client used by the staff responsible for processing the target feedback information, thereby sending the target feedback information to the first client to complete the rapid distribution of the target feedback information.
[0062] It should be noted that, when there are multiple reference tag chains, the first client identifiers corresponding to the reference tag chains may be the same or different. Therefore, when there are multiple first client identifiers corresponding to the reference tag chains, a first client identifier corresponding to a reference tag chain can be selected to send the target feedback information. Alternatively, the target feedback information can be sent to multiple first clients separately according to all the first client identifiers, which is not limited in this disclosure.
[0063] In this embodiment, the target feedback information to be processed is first obtained, and then the first label chain corresponding to the target feedback information is determined. The first label chain is then matched with each historical label chain in the historical feedback database to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain. Afterwards, the target feedback information is sent to the first client. Thus, by performing multi-level classification on the target feedback information, the first label chain is obtained, and then based on the first label chain, the client identifier corresponding to the historical label chain matching the target feedback information is obtained, thereby sending the target feedback information to the client corresponding to the client identifier to complete the distribution of the feedback information. Rapid positioning and automatic distribution of personnel who handle user feedback are achieved, which reduces the time consumption of the feedback circulation link, improves the efficiency of feedback processing, and optimizes the user experience.
[0064] Figure 2 It is a flowchart of a feedback processing method proposed in another embodiment of the present disclosure.
[0065] like Figure 2 As shown, the feedback processing method includes:
[0066] S201: Obtain target feedback information to be processed.
[0067] For the detailed description of S201 above, please refer to other embodiments of the present disclosure and will not be repeated here.
[0068] S202: Input the first prompt information and the target feedback information into the large language model respectively to obtain the first-level target label output by the large language model.
[0069] The first prompt information is used to instruct the large language model to select a first-level target tag that semantically matches the target feedback information from the first-level candidate tags.
[0070] In the disclosed embodiment, all level 1 tags in the tag database may be first obtained as first level candidate tags, and then the first prompt information may be generated according to the first level candidate tags.
[0071] For example, the first-level candidate tags are transmission, storage, account, activity, and value-added. The first prompt information may be "Among the candidate tags shown below: transmission, storage, account, activity, and value-added, filter the tags that match the target feedback information:...," etc. The present disclosure does not limit this.
[0072] It can be understood that, when the first-level candidate tags in the tag database remain unchanged, the first prompt information used to be input into the large language model to screen out the first-level target tags is also fixed.
[0073] In the disclosed embodiment, the first prompt information and the target feedback information can be respectively input into the large language model. The large language model uses its natural language analysis capability to determine which candidate tags in the first prompt information are consistent with the semantics of the target feedback information, so that the first-level target tag can be output.
[0074] It should be noted that the first-level target label output by the large language model may be one or more.
[0075] S203: Acquire the next level candidate tags associated with the first level target tags.
[0076] In the disclosed embodiment, the tag database may include association relationships between candidate tags at various levels, so the tag database may be queried for the next level candidate tags associated with each first level target tag.
[0077] S204: Generate second prompt information according to the first-level target tag and the next-level candidate tag.
[0078] The second prompt information is used to instruct the large language model to select a next-level target tag that semantically matches the target feedback information from the next-level candidate tags.
[0079] In the disclosed embodiment, the second prompt information may be generated by utilizing the first-level target tag and the association relationship between the first-level target tag and the next-level candidate tag.
[0080] For example, the first-level target tags include transmission, storage, and account, the next-level candidate tags associated with transmission include upload and backup, the next-level candidate tags associated with storage include file and decompression, and the next-level candidate tags associated with account are account ban. Then the first prompt information can be "Select the tags that match the target feedback information: ... from the candidate tags shown below: transmission, transmission-upload, transmission-backup, storage, storage-file, storage-decompression, account, account-account ban", etc., and the present disclosure does not limit this.
[0081] S205: Input the second prompt information and the target feedback information into the large language model respectively to obtain the second-level target label output by the large language model.
[0082] In the disclosed embodiment, the second prompt information and the target feedback information can be respectively input into the large language model. The large language model uses its natural language analysis capability to determine which candidate tags in the second prompt information are consistent with the semantics of the target feedback information, so that the second-level target tag can be output.
[0083] As in the above example, the second-level target labels output by the large language model may be transmission-upload, transmission-backup, and storage-file.
[0084] S206: Based on the second-level target tag, return to execute to obtain the associated next-level candidate tag, and repeat the above operation until the target tags of each level corresponding to the target feedback information are obtained.
[0085] In the disclosed embodiment, the operation of determining the next-level candidate tag associated with the second-level target tag in the tag database, i.e., S203, can be returned, and then the following S204 and S205 can be repeated to determine the next-level target tag step by step until the tag database does not contain the next-level candidate tag associated with the target tag, thereby obtaining target tags of each level corresponding to the target feedback information.
[0086] S207: Connect target labels at all levels to obtain a first label chain.
[0087] In the embodiment of the present disclosure, according to the association relationship between target tags at various levels pre-set in the tag database, target tags at various levels may be spliced in ascending order of the levels to obtain a first tag chain.
[0088] For example, the first-level target tags include transmission and storage, the second-level target tags include upload, backup, and file, and the third-level target tags include upload failure, new upload function, and backup failure. Upload and backup are the next-level tags associated with transmission, files are the next-level tags associated with storage, upload failure and new upload function are the next-level tags associated with upload, and backup failure is the next-level tag associated with backup. Then the first tag chain can include: transmission-upload-upload failure, transmission-upload-new upload function, transmission-backup-backup failure, and storage-file.
[0089] S208: Match the first label chain with each historical label chain in the historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain.
[0090] S209: Send the target feedback information to the first client.
[0091] For the detailed description of S208 and S209 above, please refer to other embodiments of the present disclosure and will not be repeated here.
[0092] In this embodiment, by using a large language model to intelligently classify the target feedback information, the efficiency of determining the first label chain is improved, which provides conditions for reducing the time consumption of automatic distribution and improving the efficiency of feedback processing.
[0093] Figure 3 It is a flowchart of a feedback processing method proposed in another embodiment of the present disclosure.
[0094] like Figure 3 As shown, the feedback processing method includes:
[0095] S301: Obtain target feedback information to be processed.
[0096] S302: Determine a first label chain corresponding to target feedback information.
[0097] S303: Match the first label chain with each historical label chain in the historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain.
[0098] For the detailed description of S301 to S303 above, please refer to other embodiments of the present disclosure and will not be repeated here.
[0099] It should be noted that in the present disclosure, after using the first label chain to match and filter the historical label chain to obtain the parameter label chain, the parameter label chain can be further screened using the parameter information of the client sending the feedback information and the similarity between the feedback information to improve the credibility of the reference label chain.
[0100] S304: Acquire first parameter information of the second client that sends target feedback information.
[0101] The second client refers to the client used by the user when inputting target feedback information in the customer service window.
[0102] The client's parameter information may include the client's platform, such as iOS, Android, PC, Web, etc., and the software version installed on the client at that time, etc., which is not limited in the present disclosure. In other words, the first parameter information is the specific platform and specific software version of the second client.
[0103] S305: Match the first parameter information with the second parameter information of the first client to obtain a first client identifier corresponding to the second parameter information matching the first parameter information.
[0104] The second parameter information of the first client refers to the parameter information corresponding to the feedback problem that the processing personnel can handle in the first client, and may also include the platform and software version, etc.
[0105] In the disclosed embodiment, when any second parameter information is identical to the first parameter information, it can be determined that the historical feedback information and the target feedback information corresponding to the reference tag chain are fed back in the client of the same platform and / or the same software version, and therefore their solutions may be the same, and the processing personnel are also the same, then the first client identifier corresponding to the matching second parameter information can be determined as the identifier of the first client to which the target feedback information is to be sent.
[0106] S306: Send the target feedback information to the first client.
[0107] For the detailed description of S306 above, please refer to other embodiments of the present disclosure and will not be repeated here.
[0108] In this embodiment, the first client identifier corresponding to the filtered reference tag chain is further filtered by using the client's parameter information, thereby further improving the reliability of the first client for target feedback information distribution and providing conditions for improving the efficiency of feedback processing.
[0109] Figure 4 It is a flowchart of a feedback processing method proposed in another embodiment of the present disclosure.
[0110] like Figure 4 As shown, the feedback processing method includes:
[0111] S401: Obtain target feedback information to be processed.
[0112] S402: Determine a first label chain corresponding to target feedback information.
[0113] S403: Match the first label chain with each historical label chain in the historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain.
[0114] For the detailed description of S401 to S403 above, please refer to other embodiments of the present disclosure and will not be repeated here.
[0115] S404: When there are multiple first clients, determine the similarity between the target feedback information and the historical feedback information corresponding to the reference tag chain.
[0116] In the disclosed embodiment, when the first client identifiers corresponding to any two reference label chains are different, it can be determined that there are multiple first clients. Then, the target feedback information can be summarized using a large language model to obtain a summary text, and then the summary text can be vectorized to obtain a vector corresponding to the target feedback information. After that, the vector corresponding to the target feedback information can be respectively compared with the vector of the historical feedback information corresponding to each reference label chain, and the cosine distance can be calculated to obtain the similarity between the target feedback information and the historical feedback information corresponding to the reference label chain.
[0117] Alternatively, the target feedback information and the historical feedback information may be directly matched to obtain similarity, etc., which is not limited in the present disclosure.
[0118] S405: Send the target feedback information to the first client corresponding to the reference tag chain with the highest similarity.
[0119] In the disclosed embodiment, when any historical feedback information has the highest similarity with the target feedback information, it can be determined that the processing scheme of the historical feedback information is more suitable for the target feedback information, and the first client corresponding to the reference tag chain with the highest similarity can be determined as the first client to which the target feedback information is sent.
[0120] It should be noted that there may be two reference label chains with the same similarity. Therefore, when there is more than one reference label chain with the highest similarity and the first client identifier corresponding to each reference label chain is also different, other conditions are needed to further sort the priorities of the two reference label chains.
[0121] Optionally, when the number of reference label chains with the highest similarity is greater than 1, the levels of the highest-level labels that match the first label chain respectively with the multiple reference label chains can be determined. Then, according to the level, the target feedback information is sent to the first client corresponding to the reference label chain with the largest level.
[0122] For example, the reference label chains with the highest similarity are reference label chain A and reference label chain B, and the first client identifier corresponding to reference label chain A is a, and the first client identifier corresponding to reference label chain B is b. It is determined that the level of the highest-level label that matches reference label chain A and the first label chain is 4, and the level of the highest-level label that matches reference label chain B and the first label chain is 3. Since 4>3, it can be determined that the target feedback information is sent to the first client a.
[0123] In the disclosed embodiment, when the similarities of multiple reference label chains are the same, the levels of the highest-level labels that match the reference label chains and the first label chain are compared to determine the first client to send a target feedback information, thereby further improving the reliability and efficiency of feedback distribution.
[0124] It should be noted that, in the present disclosure, in addition to sending the target feedback information to the first client, a reference processing solution can be provided to the processing personnel based on the historical processing data of the historical feedback, thereby further improving the efficiency of the feedback processing.
[0125] Optionally, processing data of historical feedback information corresponding to the reference tag chain may be obtained, and then the processing data and target feedback information may be sent to the first client.
[0126] The processing data of historical feedback information may include the reproduction path, problem attribution, and solution, etc.
[0127] In the disclosed embodiment, the processing data of the historical feedback information corresponding to all reference tag chains can be sent to the first client together with the target feedback information to provide a processing solution for reference to the feedback processing personnel, further improve the efficiency of feedback processing, and reduce the workload of the feedback processing personnel.
[0128] It should be noted that since there can be multiple reference tag chains, there may also be multiple processing data of the historical feedback information corresponding to the reference tag chain. If all the processing data is sent to the first client without processing, it will not provide the feedback processing personnel with a good processing solution reference, which may cause confusion to the processing personnel and affect the processing efficiency.
[0129] Optionally, when there are multiple processed data, the weight value corresponding to each reference label chain can be determined based on the similarity corresponding to each reference label chain, and then the multiple processed data and target feedback information can be sent to the first client in sequence according to the weight value.
[0130] The weight value corresponding to the reference label chain is used to represent the credibility of the processed data corresponding to the reference label chain when processing the target feedback information.
[0131] In the disclosed embodiment, the similarity corresponding to each reference label chain can be determined based on the similarity between the target feedback information and the historical feedback information corresponding to the reference label chain, and then the weight value corresponding to each reference label chain can be determined based on the magnitude of the similarity. The greater the similarity, the greater the weight value. After obtaining the weight value corresponding to each reference label chain, the corresponding processed data is sorted in descending order of the weight value, and the sorted multiple processed data are sent to the first client together with the target feedback information.
[0132] It should be noted that, in the present disclosure, a similarity threshold may also be set. When the similarity corresponding to any reference tag chain is less than the similarity threshold, the processed data corresponding to the reference tag chain is deleted and not sent to the first client.
[0133] In the disclosed embodiment, by determining the similarity corresponding to each reference tag chain, the order in which the processed data is sent to the first client is sorted, so that the data sent to the first client is more organized, which facilitates the processing personnel to quickly and accurately obtain the reference processing solution, and further improves the efficiency of feedback processing.
[0134] In this embodiment, the most reliable first client is determined among multiple first client identifiers based on the similarity between the historical feedback information corresponding to the reference tag chain and the target feedback information, and the target feedback information is sent, thereby further improving the accuracy and reliability of feedback distribution and improving the efficiency of feedback processing.
[0135] Figure 5 It is a structural diagram of a feedback processing device proposed in an embodiment of the present disclosure.
[0136] like Figure 5 As shown, the feedback processing device 500 includes:
[0137] An acquisition module 501 is used to acquire target feedback information to be processed;
[0138] A first determination module 502, configured to determine a first label chain corresponding to target feedback information;
[0139] The second determination module 503 is used to match the first label chain with each historical label chain in the historical feedback database respectively, to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain;
[0140] The sending module 504 is used to send the target feedback information to the first client.
[0141] Optionally, the first determining module 502 may be specifically configured to:
[0142] Obtain multi-level candidate labels and the association relationship between candidate labels;
[0143] Match the target feedback information with the candidate labels at each level, and determine the labels at each level that match the target feedback information;
[0144] According to the association relationship between candidate labels at each level, labels at each level that match the target feedback information are concatenated to obtain a first label chain.
[0145] Optionally, the first determining module 502 may be specifically configured to:
[0146] Inputting the first prompt information and the target feedback information into the large language model respectively to obtain the first-level target label output by the large language model;
[0147] Get the next level candidate labels associated with the first level target labels;
[0148] Generate second prompt information according to the first-level target label and the next-level candidate label;
[0149] Inputting the second prompt information and the target feedback information into the large language model respectively to obtain the second-level target label output by the large language model;
[0150] Based on the second-level target label, return to execute and obtain the associated next-level candidate label, and repeat the above operation until the target labels of each level corresponding to the target feedback information are obtained;
[0151] The target labels at all levels are concatenated to obtain the first label chain.
[0152] Optionally, the second determining module 503 may also be used to:
[0153] Acquire first parameter information of a second client that sends target feedback information;
[0154] The first parameter information is matched with the second parameter information of the first client to obtain a first client identifier corresponding to the second parameter information matching the first parameter information.
[0155] Optionally, the sending module 504 may be specifically configured to:
[0156] In the case where there are multiple first clients, determining the similarity between the target feedback information and the historical feedback information corresponding to the reference tag chain;
[0157] The target feedback information is sent to the first client corresponding to the reference tag chain with the highest similarity.
[0158] Optionally, the sending module 504 may be specifically configured to:
[0159] When the number of reference label chains with the highest similarity is greater than 1, determining the levels of the highest-level labels that the multiple reference label chains respectively match with the first label chain;
[0160] According to the size of the level, the target feedback information is sent to the first client corresponding to the reference tag chain with the largest level.
[0161] Optionally, the sending module 504 may be specifically configured to:
[0162] Obtain processing data of historical feedback information corresponding to the reference tag chain;
[0163] The processed data and target feedback information are sent to the first client.
[0164] Optionally, the sending module 504 may be specifically configured to:
[0165] When there are multiple processing data, the weight value corresponding to each reference label chain is determined based on the similarity corresponding to each reference label chain;
[0166] According to the weight values, the plurality of processed data and the target feedback information are sequentially sent to the first client.
[0167] It should be noted that the above explanation of the feedback processing method is also applicable to the feedback processing device of this embodiment and will not be repeated here.
[0168] In this embodiment, the target feedback information to be processed is first obtained, and then the first label chain corresponding to the target feedback information is determined. The first label chain is then matched with each historical label chain in the historical feedback database to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain. Afterwards, the target feedback information is sent to the first client. Thus, by performing multi-level classification on the target feedback information, the first label chain is obtained, and then based on the first label chain, the client identifier corresponding to the historical label chain matching the target feedback information is obtained, thereby sending the target feedback information to the client corresponding to the client identifier to complete the distribution of the feedback information. Rapid positioning and automatic distribution of personnel who handle user feedback are achieved, which reduces the time consumption of the feedback circulation link, improves the efficiency of feedback processing, and optimizes the user experience.
[0169] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
[0170] Figure 6A schematic block diagram of an example electronic device 600 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.
[0171] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.
[0172] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0173] The computing unit 601 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 601 performs the various methods and processes described above, such as the topic generation method. For example, in some embodiments, the topic generation method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the topic generation method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the topic generation method in any other appropriate manner (e.g., by means of firmware).
[0174] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0175] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0176] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0177] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).
[0178] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.
[0179] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.
[0180] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.
[0181] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. In the description of the present disclosure, the words "if" and "if" used can be interpreted as "at the time of" or "when" or "in response to determination" or "under the circumstances of".
[0182] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.
Claims
1. A feedback processing method, It is characterized in that include: Obtain target feedback information to be processed; Determine a first label chain corresponding to the target feedback information; Matching the first label chain with each historical label chain in a historical feedback database respectively to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain; The target feedback information is sent to the first client.
2. The method according to claim 1, It is characterized in that The determining a first label chain corresponding to the target feedback information includes: Obtain multi-level candidate labels and the association relationship between candidate labels; Matching the target feedback information with each level of candidate labels to determine each level of labels that match the target feedback information; According to the association relationship between the candidate labels at each level, the labels at each level that match the target feedback information are spliced to obtain the first label chain.
3. The method according to claim 1, wherein determining the first label chain corresponding to the target feedback information, include: Inputting the first prompt information and the target feedback information into the large language model respectively to obtain a first-level target label output by the large language model; Obtaining a next-level candidate tag associated with the first-level target tag; Generate second prompt information according to the first-level target label and the next-level candidate label; Inputting the second prompt information and the target feedback information into the large language model respectively to obtain a second-level target label output by the large language model; Based on the second-level target tag, return to execute the acquisition of the associated next-level candidate tag, and repeat the above operation until the target tags of each level corresponding to the target feedback information are obtained; The target labels at each level are concatenated to obtain the first label chain.
4. The method according to claim 1, It is characterized in that After obtaining a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain, the method further includes: Acquire first parameter information of a second client that sends the target feedback information; The first parameter information is matched with the second parameter information of the first client to obtain a first client identifier corresponding to the second parameter information matching the first parameter information.
5. The method according to claim 1, It is characterized in that The sending the target feedback information to the first client includes: In the case that there are multiple first clients, determining the similarity between the target feedback information and the historical feedback information corresponding to the reference tag chain; The target feedback information is sent to the first client corresponding to the reference tag chain with the highest similarity.
6. The method according to claim 5, It is characterized in that The sending the target feedback information to the first client corresponding to the reference tag chain with the highest similarity includes: When the number of reference label chains with the highest similarity is greater than 1, determining the levels of the highest-level labels that the multiple reference label chains respectively match with the first label chain; According to the size of the level, the target feedback information is sent to the first client corresponding to the reference tag chain with the largest level.
7. The method according to any one of claims 5-6, It is characterized in that The sending the target feedback information to the first client includes: Acquire processing data of historical feedback information corresponding to the reference tag chain; The processed data and the target feedback information are sent to the first client.
8. The method according to claim 7, It is characterized in that The sending the processed data and the target feedback information to the first client includes: In the case where there are multiple processed data, determining a weight value corresponding to each reference label chain based on the similarity corresponding to each reference label chain; According to the weight value, the plurality of processed data and the target feedback information are sequentially sent to the first client.
9. A feedback processing device, It is characterized in that include: An acquisition module, used to acquire target feedback information to be processed; A first determining module, used to determine a first label chain corresponding to the target feedback information; A second determination module is used to match the first label chain with each historical label chain in a historical feedback database respectively, to obtain a reference label chain matching the first label chain and a first client identifier corresponding to the reference label chain; A sending module is used to send the target feedback information to the first client.
10. The device according to claim 9, It is characterized in that The first determining module is specifically configured to: Obtain multi-level candidate labels and the association relationship between candidate labels; Matching the target feedback information with each level of candidate labels to determine each level of labels that match the target feedback information; According to the association relationship between the candidate labels at each level, the labels at each level that match the target feedback information are spliced to obtain the first label chain.
11. The device according to claim 9, wherein the first determining module is specifically configured to: Inputting the first prompt information and the target feedback information into the large language model respectively to obtain a first-level target label output by the large language model; Obtaining a next-level candidate tag associated with the first-level target tag; Generate second prompt information according to the first-level target label and the next-level candidate label; Inputting the second prompt information and the target feedback information into the large language model respectively to obtain a second-level target label output by the large language model; Based on the second-level target tag, return to execute the acquisition of the associated next-level candidate tag, and repeat the above operation until the target tags of each level corresponding to the target feedback information are obtained; The target labels at each level are concatenated to obtain the first label chain.
12. The device according to claim 9, It is characterized in that The second determining module is further used for: Acquire first parameter information of a second client that sends the target feedback information; The first parameter information is matched with the second parameter information of the first client to obtain a first client identifier corresponding to the second parameter information matching the first parameter information.
13. The device according to claim 9, It is characterized in that The sending module is specifically used for: In the case that there are multiple first clients, determining the similarity between the target feedback information and the historical feedback information corresponding to the reference tag chain; The target feedback information is sent to the first client corresponding to the reference tag chain with the highest similarity.
14. The device according to claim 13, It is characterized in that The sending module is specifically used for: When the number of reference label chains with the highest similarity is greater than 1, determining the levels of the highest-level labels that the multiple reference label chains respectively match with the first label chain; According to the size of the level, the target feedback information is sent to the first client corresponding to the reference tag chain with the largest level.
15. The device according to any one of claims 13 to 14, It is characterized in that The sending module is specifically used for: Acquire processing data of historical feedback information corresponding to the reference tag chain; The processed data and the target feedback information are sent to the first client.
16. The device according to claim 15, It is characterized in that The sending module is specifically used for: In the case where there are multiple processed data, determining a weight value corresponding to each reference label chain based on the similarity corresponding to each reference label chain; According to the weight value, the plurality of processed data and the target feedback information are sequentially sent to the first client.
17. An electronic device, include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the feedback processing method according to any one of claims 1 to 8.
18. A non-transitory computer-readable storage medium storing computer instructions, It is characterized in that in, The computer instructions are used to enable the computer to execute the feedback processing method according to any one of claims 1 to 8.
19. A computer program product, It is characterized in that The method comprises a computer program, which, when executed by a processor, implements the steps of the feedback processing method according to any one of claims 1 to 8.
Citation Information
Patent Citations
Feedback assignment method and system based on semantic analysis
CN107301229A
Data processing method and device, computer equipment and storage medium
CN109783697A
Database construction method and system
CN111209491A
Information management method and system based on big data and Internet, and cloud server
CN112765386A
Work order processing method and device, computer equipment and storage medium
CN115730925A
