Information distribution method, system, equipment and medium

Abstract

The invention relates to an information distribution method. The method comprises the following steps: receiving feedback information of an object; determining reference classification information of each service system, and determining the reference classification information matched with the feedback information as target classification information; and sending the feedback information to a service system corresponding to the target classification information. A classification list of the feedback information does not need to be provided for the user, the classification difficulty of the feedback information of the user is effectively reduced, the user experience is improved, and meanwhile, the classification accuracy of the feedback information is further improved compared with understanding classification of the user according to the reference classification information of each service system.

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular to an information distribution method, system, device, and medium. Background Art

[0002] With the development of the internet, internet products have entered the era of inventory. Efficient and refined product operations and improved user experience are of paramount importance. To enhance this experience, an increasing number of internet products have internal user feedback collection systems to receive user feedback, such as failures and suggestions. Furthermore, this feedback is distributed to the appropriate internal business departments for processing.

[0003] However, most traditional user feedback information collection systems require users to first select a corresponding category from a classification list and then fill in the feedback content before submitting. These categories are generally set to correspond to the work content of internal business departments. Since users cannot know the specific content of each business department, they may have misunderstandings when providing feedback based on the provided classification list, resulting in an inability to accurately select the corresponding category, which in turn causes the corresponding feedback information to be sent to the wrong business department for processing. In some cases, some users even become dissatisfied because they cannot find the classification list corresponding to their feedback content, and they give up on providing feedback related issues or suggestions.

[0004] Therefore, the traditional feedback information collection system rigidly "passes" the problem of accurate classification of feedback information to the user, which not only affects the accuracy of feedback information delivery, but may also cause the feedback information to be "lost" due to the user's emotions when providing feedback. Summary of the Invention

[0005] The present application provides an information distribution method to solve the problems of inaccurate user feedback information distribution and poor user experience in traditional feedback information collection systems.

[0006] In a first aspect, the present application provides an information distribution method, comprising:

[0007] Receive feedback information from the object;

[0008] Determining reference classification information for each business system, and determining reference classification information that matches the feedback information as target classification information, wherein the reference classification information is character information determined based on a functional description of a business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information;

[0009] The feedback information is sent to the business system corresponding to the target classification information.

[0010] In some embodiments of the present application, the reference classification information includes: service description information;

[0011] The determining of reference classification information matching the feedback information as target classification information includes:

[0012] Determine the business description information corresponding to each business system, and obtain a set of business description information corresponding to all business systems;

[0013] Inputting the service description information in the service description information set and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the service description information set;

[0014] The target service description information is used as the target classification information.

[0015] In some embodiments of the present application, the reference classification information includes historical feedback information, where the historical feedback information is feedback information that has been historically processed by the business system;

[0016] The determining of reference classification information matching the feedback information as target classification information includes:

[0017] Determine the corresponding historical feedback information in each business system and obtain the historical feedback information set corresponding to all business systems;

[0018] Determining text similarity between the feedback information and any historical feedback information in the historical feedback information set;

[0019] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0020] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0021] If the number of historical feedback information whose text similarity meets the first predetermined condition is greater than a predetermined value, the business system corresponding to the historical feedback information whose text similarity meets the first predetermined condition is selected as a candidate business system;

[0022] Taking the service description information corresponding to the to-be-selected service system as the to-be-selected service description information, and obtaining a to-be-selected service description information set;

[0023] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0024] The target service description information is used as the target classification information.

[0025] In some embodiments of the present application, the reference classification information includes classification keywords, which are characters contained in feedback information that has been historically processed by the corresponding business system;

[0026] The determining of reference classification information matching the feedback information as target classification information includes:

[0027] Segmenting the feedback information to obtain a corresponding keyword set;

[0028] The classification keywords included in the keyword set are used as target classification information that matches the feedback information.

[0029] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0030] If the number of business systems corresponding to the classification keywords included in the keyword set is greater than a predetermined value, the business systems corresponding to the classification keywords included in the keyword set are used as candidate business systems;

[0031] Determine the corresponding historical feedback information in each candidate business system, and obtain a set of candidate historical feedback information corresponding to all candidate business systems;

[0032] Determining text similarity between the feedback information and any historical feedback information in the set of to-be-selected historical feedback information;

[0033] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0034] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0035] If the number of business systems corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, determining the business description information corresponding to the candidate business system, and obtaining a set of candidate business description information corresponding to all candidate business systems;

[0036] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0037] The target service description information is used as the target classification information.

[0038] A second aspect of the present application provides an information distribution system, comprising:

[0039] An information receiving module, used for receiving feedback information from the object;

[0040] An information classification module is used to determine reference classification information of each business system, and based on a matching strategy corresponding to the reference classification information, determine the reference classification information that matches the feedback information as the target classification information;

[0041] The information distribution module is used to send the feedback information to the business system corresponding to the target classification information.

[0042] The third aspect of the present application also proposes a computer device: at least one communication interface; at least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; at least one memory connected to the at least one bus, wherein the processor is configured to execute the information distribution method described in any one of the above embodiments.

[0043] The fourth aspect of the present application further proposes a computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the information distribution method described in any one of the above embodiments.

[0044] The technical solution provided by the embodiments of the present application has the following advantages over existing technologies: The method provided by the embodiments of the present application directly receives user feedback information and, after receiving the feedback information, determines the business system to which the feedback information should be distributed based on the corresponding reference classification information of each business system and distributes it to that system. This eliminates the need to provide users with a categorized list of feedback information, effectively reducing the difficulty of categorizing user feedback information and improving the user experience. Furthermore, based on the reference classification information of each business system, the accuracy of feedback information classification is further improved compared to user-understood classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings. These exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0048] Figure 1 A flowchart of an information distribution method provided in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a process for determining target classification information based on business description information of a business system provided in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of a process for determining target classification information based on historical feedback information of a business system provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of a process for determining target classification information based on service description information corresponding to historical feedback information provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram of a process for determining target classification information based on assigned keywords provided in an embodiment of the present application;

[0053] Figure 6 A schematic diagram of a process for determining target classification information based on classification keywords and historical feedback information provided in an embodiment of the present application;

[0054] Figure 7 A schematic diagram of a process for determining target classification information based on the business description information corresponding to the classification keyword provided in an embodiment of the present application;

[0055] Figure 8 A schematic diagram of the structure of an information distribution system provided in an embodiment of the present application;

[0056] Figure 9 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0058] The disclosure below provides many different embodiments or examples for implementing different configurations of the present invention. To simplify the disclosure of the present invention, the components and configurations of specific examples are described below. Of course, these are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numerals and / or letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or configurations discussed.

[0059] When collecting user feedback, traditional feedback collection systems first need to provide users with corresponding classification options. Users must select the corresponding classification option and then enter the content they want to feedback. The feedback collection system sends the user's feedback information to the corresponding business system based on the classification options. The classification options correspond to the business system that processes the classified feedback information. Whether the feedback information is accurately sent to the business system that handles the reported problem depends on the user's understanding of the classification options. This method leads to the unreliability of feedback information distribution. In addition, for some users, the cumbersome classification options exacerbate the user's negative emotions when they encounter the negative impact of related problems, which may cause the user to give up on reporting related problems, and ultimately make the business system unable to obtain relevant problems for improvement.

[0060] like Figure 1 As shown, in order to solve the above problems, in a first aspect, the present application provides an information distribution method, comprising:

[0061] Step S1, receiving feedback information from the object;

[0062] Step S2: Determine reference classification information for each business system, and determine reference classification information that matches the feedback information as target classification information, wherein the reference classification information is character information determined based on a functional description of a business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information;

[0063] Step S3: Send the feedback information to the business system corresponding to the target classification information.

[0064] In the embodiments of the present application, the objects are relevant users who provide feedback information or other information collection systems that forward feedback information on their behalf (because, in some cases, there may be scenarios where user feedback information is collected by a third party).

[0065] The business system refers to a system for receiving feedback information corresponding to the business department that handles the problem corresponding to a certain type of feedback information. In some embodiments, it refers to a data interface in the form of corresponding software or a data storage interface or storage space.

[0066] Reference classification information refers to the character information used to represent the corresponding type of feedback information processed by each business system. It is used to distinguish the feedback information processed by different business systems. It can be a functional description of each business system, or a business description and business scope description.

[0067] In step S1, feedback information sent by the user is received. The feedback information may not contain any labels related to the classification options, but only the problem description entered by the user and corresponding problem evidence, such as pictures, screenshots, etc.

[0068] In step S2, the business system that will process the feedback information is first determined. Based on the determined business system, the corresponding reference classification information for each business system is determined. For example, the current business description of each business system is used. After determining the reference classification information for each system, the content of the received feedback information is analyzed to determine which reference classification information is most relevant. The most relevant reference classification information is then used as the target classification information.

[0069] In step S3, the received feedback information is sent to the business system corresponding to the target classification information determined in step S2.

[0070] Specifically, in one application scenario, the feedback information collection system serves a video content provider, and the video content provider has multiple business systems inside.

[0071] For example, business system a is used to handle user video content production, video upload, work review and other related businesses;

[0072] Business system b, used to process user purchase, payment, activation, redemption, upgrade of user membership and other membership-related services;

[0073] Business system c, used to handle membership rights and interests related to gold, platinum, and diamond membership, such as validity period and points;

[0074] Business system d, used to handle advertising display-related services such as full-screen advertising, patch advertising, and advertising duration;

[0075] It also includes a special business system e, which is used to process other content that does not belong to the reference classification information corresponding to the above business systems.

[0076] Furthermore, after the feedback information collects the corresponding feedback information from the user, the corresponding target classification information is determined based on the content of the feedback information. For example, a user submits a feedback message, which contains the user's description of the problem encountered. For example, the user describes that after recharging the membership, he did not obtain the corresponding membership rights, resulting in the inability to play the video content he wanted to watch, and took a screenshot as evidence. For this feedback information, combined with the reference classification information of each of the above-mentioned business systems, it can be seen that the closest to the problem reported by the user should be the corresponding reference classification information business of business system b, that is, "purchase, payment, activation, redemption, upgrade of user membership and other membership purchases", and the reference classification information of business system b is used as the target classification information.

[0077] Furthermore, the feedback information collection system sends the feedback information to the business system b, thus completing the distribution operation of the user information.

[0078] like Figure 2 As shown, in some embodiments of the present application, the reference classification information includes: business description information;

[0079] The determining of reference classification information matching the feedback information as target classification information includes:

[0080] Step S21: Determine the business description information corresponding to each business system, and obtain a set of business description information corresponding to all business systems;

[0081] Step S22: inputting the service description information in the service description information set and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the service description information set;

[0082] Step S23: Use the target service description information as the target classification information.

[0083] In this embodiment, the large language model refers to an artificial intelligence model with text processing capabilities, or a multimodal model that can realize multimodal recognition conversion such as image-text conversion, and is used to process the image and text content in the user's feedback information.

[0084] The business description information is the comprehensive description of the business functions and business scope of each business system in the aforementioned embodiments. For example, business system A includes "production, uploading, and review of video works"; business system B includes "purchase, payment, activation, redemption, and upgrade of user memberships," etc.

[0085] Specifically, the business description information of each business system is determined, and then a business description information set is constructed based on the business description information of each business system.

[0086] Furthermore, the service description information and feedback information in the service description information set are input into the large model, and the large model determines the service description information in the service description information set that is most similar to the feedback information. The service description information determined by the large model is then used as target classification information for subsequent classification of the feedback information.

[0087] In some embodiments of the present application, when the service description information and feedback information in the service description information set are input into the macro model, a corresponding prompt word project is determined based on each service description information. The prompt word project corresponding to each service description information is also input into the macro model, providing the macro model with a reference case for matching the feedback information and the service description information, thereby producing a more reliable classification result.

[0088] like Figure 3 As shown, in some embodiments of the present application, the reference classification information includes historical feedback information, and the historical feedback information is feedback information that has been historically processed by the business system;

[0089] The determining of reference classification information matching the feedback information as target classification information includes:

[0090] Step S31: determine the corresponding historical feedback information in each business system, and obtain a set of historical feedback information corresponding to all business systems;

[0091] Step S31: determining the text similarity between the feedback information and any historical feedback information in the historical feedback information set;

[0092] Step S33: Use historical feedback information whose text similarity meets the first predetermined condition as target classification information.

[0093] In this embodiment, the reference classification information may be historical feedback information, and the historical feedback information is feedback information that has been processed by the corresponding business system.

[0094] Specifically, historical feedback information corresponding to each business system is obtained to obtain a set of historical feedback information corresponding to all business systems.

[0095] Furthermore, in some embodiments of the present application, the text similarity between the historical feedback information and the feedback information to be determined is determined by the literal representation of the feedback information. That is, the historical feedback information in the corresponding historical feedback information set is subjected to text processing and stored in the corresponding database. For example, all historical feedback information is segmented, and the segmentation results are stored in the elasticsearch database. After receiving new feedback information, the corresponding feedback information is segmented, and then the segmentation results are queried in elasticsearch to obtain one or more query results. The elasticsearch query result is the text similarity between the corresponding historical feedback information and the feedback information. If the elasticsearch database contains multiple historical feedback information, in principle the elasticsearch database will calculate the text similarity between the feedback information and each historical feedback information. In this embodiment, the result with a text similarity reaching a certain value is selected as the calculation result of the text similarity, and then the historical feedback information with the highest text similarity is selected as the target classification information.

[0096] In some embodiments of the present application, when determining the textual similarity between feedback information and any historical feedback information in a set of historical feedback information, this is achieved through the semantic representation of the feedback information and the historical feedback information. Specifically, the historical feedback information is first processed using a corresponding artificial intelligence model, such as OpenAI's text-embedding tool or open-source models such as bge-large-zh and gte-large-zh, to obtain a corresponding semantic vector. The corresponding semantic vector is then stored in a vector database, such as Chrome or Qdrant.

[0097] Furthermore, after receiving new feedback, the semantic conversion tool is used to convert the feedback into a semantic vector. Based on the semantic vector of the feedback, the semantic database corresponding to the previous feedback is used to query the vector distance between the feedback and the previous feedback, which is used as the text similarity. The historical feedback with the closest vector distance is also selected as the target classification information.

[0098] like Figure 4 As shown, in some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0099] Step S41: If the number of historical feedback information whose text similarity meets the first predetermined condition is greater than a predetermined value, the business system corresponding to the historical feedback information whose text similarity meets the first predetermined condition is selected as a candidate business system;

[0100] Step S42: Using the service description information corresponding to the to-be-selected service system as the to-be-selected service description information to obtain a to-be-selected service description information set;

[0101] Step S43: inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0102] Step S44: Use the target service description information as the target classification information.

[0103] In some cases, when matching based on text similarity, there may be multiple historical feedback pieces with the highest textual similarity. This means that two or more historical feedback pieces share the same textual similarity with the feedback piece. This means that regardless of the matching method based on the literal representation of the word segmentation or the semantic vector representation, multiple results with the same similarity or semantic vector may be predetermined. In some cases, the business systems corresponding to these multiple results are not uniform. As a result, there may be multiple results for the target classification information, making it impossible to accurately assign them to the corresponding business system.

[0104] To this end, in this embodiment, the business systems corresponding to the historical feedback information of multiple results matched by text similarity are selected as candidate business systems. The business description information corresponding to the candidate business systems is then determined. If the multiple results matched by text similarity include all business systems, the candidate business systems are all business systems.

[0105] Furthermore, as described in the previous example, a corresponding set of candidate service description information is constructed based on the candidate service system, and the service description information and feedback information in the set of candidate service description information are input into the big model, which determines which service description information matches the feedback information.

[0106] And use the matched business description information as target classification information.

[0107] like Figure 5 As shown, in some embodiments of the present application, the reference classification information includes classification keywords, and the classification keywords are characters contained in the feedback information that has been historically processed by the corresponding business system;

[0108] The determining of reference classification information matching the feedback information as target classification information includes:

[0109] Step S51: Segment the feedback information to obtain a corresponding keyword set;

[0110] Step S52: Use the category keywords included in the keyword set as target category information that matches the feedback information.

[0111] In this embodiment, the reference classification information also includes classification keywords. Each business system has a set of classification keywords associated with it. The classification keywords are characters contained in the feedback information that each business system has historically processed. In other words, the classification keywords of each business system can be found in the sand and gravel feedback information that the corresponding business system has processed.

[0112] Specifically, after receiving the feedback information, the feedback information is segmented to obtain a keyword set for the feedback information. The segmented words in the keyword set are then traversed to determine whether each segmented word is identical to the corresponding classification keyword. If they are identical, the corresponding classification keyword is used as the target classification information. When the feedback information is subsequently distributed, it can be distributed to the business system corresponding to the classification keyword.

[0113] In some embodiments of the present application, the classification keywords may be identified by frequently appearing in the historical feedback information of each business system and mutually exclusive with other business systems. That is, the keywords must appear frequently in the historical feedback information and / or not appear in the historical feedback information of other business systems.

[0114] like Figure 6 As shown, in some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0115] Step S61: If the number of business systems corresponding to the classification keywords included in the keyword set is greater than a predetermined value, the business systems corresponding to the classification keywords included in the keyword set are selected as candidate business systems;

[0116] Step S62: Determine the corresponding historical feedback information in each candidate business system, and obtain a set of candidate historical feedback information corresponding to all candidate business systems;

[0117] Step S63: Determine the text similarity between the feedback information and any historical feedback information in the set of selected historical feedback information;

[0118] Step S64: Use historical feedback information whose text similarity meets the first predetermined condition as target classification information.

[0119] In this embodiment, as described above, if the feedback information is segmented and the classification keywords are matched to determine that the feedback information contains classification keywords corresponding to multiple business systems, that is, the segmented words in the feedback information may involve classification keywords corresponding to multiple business systems, it is impossible to accurately determine the corresponding target classification information.

[0120] To this end, the business system corresponding to the matched classification keyword can be selected as the candidate business system. The target classification information is then re-determined by matching the text similarity of the historical feedback information corresponding to the candidate business system. As previously described, the feedback information is segmented or semantic vectors are calculated, and the corresponding text similarity database is queried to obtain the historical feedback information that meets the text similarity conditions, which is used as the target classification information.

[0121] like Figure 7 As shown, in some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0122] Step S71: If the number of business systems corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, then determining the business description information corresponding to the candidate business systems, and obtaining a set of candidate business description information corresponding to all candidate business systems;

[0123] Step S72: inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0124] Step S73: Use the target service description information as the target classification information.

[0125] In this embodiment, as described above, if there are still multiple historical feedback information with the same text similarity through text similarity matching, and they all correspond to different business systems.

[0126] In this case, the business description information corresponding to the candidate business systems determined by the classification keywords can be used to construct a candidate business description information set. The business description information and feedback information in the candidate business description information set are then input into the macro model. The macro model then outputs the business description information that matches the feedback information, i.e., the target business description information, and uses the target business description information as the target classification information.

[0127] The technical solution provided by the embodiments of the present application has the following advantages over existing technologies: The method provided by the embodiments of the present application directly receives user feedback information and, after receiving the feedback information, determines the business system to which the feedback information should be distributed based on the corresponding reference classification information of each business system and distributes it to that system. This eliminates the need to provide users with a categorized list of feedback information, effectively reducing the difficulty of categorizing user feedback information and improving the user experience. Furthermore, based on the reference classification information of each business system, the accuracy of feedback information classification is further improved compared to user-understood classification.

[0128] Furthermore, as shown above, keyword matching can quickly categorize feedback information. If the keyword matching result is not unique, text similarity matching can be used for further rapid matching. If the corresponding text similarity result is not unique, a large model is used to determine the final reliable result. This hierarchical screening method balances efficiency with accuracy and reliability.

[0129] like Figure 8 As shown, the second aspect of the present application provides an information distribution system, comprising:

[0130] Information receiving module 1, used to receive feedback information from the object;

[0131] Information classification module 2 is configured to determine reference classification information for each business system and, based on a matching strategy corresponding to the reference classification information, determine reference classification information that matches the feedback information as target classification information. The reference classification information is character information determined based on a functional description of the business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information.

[0132] The information distribution module 3 is used to send the feedback information to the business system corresponding to the target classification information.

[0133] like Figure 9 As shown, the embodiment of the present application provides a computer device, including a processor 911, a communication interface 912, a memory 913 and a communication bus 914, wherein the processor 911, the communication interface 912, and the memory 913 communicate with each other through the communication bus 914.

[0134] Memory 913, for storing computer programs;

[0135] In one embodiment of the present application, the processor 911 is configured to, when executing a program stored in the memory 913, implement any one of the aforementioned method embodiments to provide an information distribution method, including:

[0136] Receive feedback information from the object;

[0137] Determining reference classification information for each business system, and determining reference classification information that matches the feedback information as target classification information, wherein the reference classification information is character information determined based on a functional description of a business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information;

[0138] The feedback information is sent to the business system corresponding to the target classification information.

[0139] In some embodiments of the present application, the reference classification information includes: service description information;

[0140] The determining of reference classification information matching the feedback information as target classification information includes:

[0141] Determine the business description information corresponding to each business system, and obtain a set of business description information corresponding to all business systems;

[0142] Inputting the service description information in the service description information set and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the service description information set;

[0143] The target service description information is used as the target classification information.

[0144] In some embodiments of the present application, the reference classification information includes historical feedback information, where the historical feedback information is feedback information that has been historically processed by the business system;

[0145] The determining of reference classification information matching the feedback information as target classification information includes:

[0146] Determine the corresponding historical feedback information in each business system and obtain the historical feedback information set corresponding to all business systems;

[0147] Determining text similarity between the feedback information and any historical feedback information in the historical feedback information set;

[0148] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0149] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0150] If the number of historical feedback information whose text similarity meets the first predetermined condition is greater than a predetermined value, the business system corresponding to the historical feedback information whose text similarity meets the first predetermined condition is selected as a candidate business system;

[0151] Taking the service description information corresponding to the to-be-selected service system as the to-be-selected service description information, and obtaining a to-be-selected service description information set;

[0152] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0153] The target service description information is used as the target classification information.

[0154] In some embodiments of the present application, the reference classification information includes classification keywords, which are characters contained in feedback information that has been historically processed by the corresponding business system;

[0155] The determining of reference classification information matching the feedback information as target classification information includes:

[0156] Segmenting the feedback information to obtain a corresponding keyword set;

[0157] The classification keywords included in the keyword set are used as target classification information that matches the feedback information.

[0158] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0159] If the number of business systems corresponding to the classification keywords included in the keyword set is greater than a predetermined value, the business systems corresponding to the classification keywords included in the keyword set are used as candidate business systems;

[0160] Determine the corresponding historical feedback information in each candidate business system, and obtain a set of candidate historical feedback information corresponding to all candidate business systems;

[0161] Determining text similarity between the feedback information and any historical feedback information in the set of to-be-selected historical feedback information;

[0162] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0163] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0164] If the number of business systems corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, determining the business description information corresponding to the candidate business system, and obtaining a set of candidate business description information corresponding to all candidate business systems;

[0165] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0166] The target service description information is used as the target classification information.

[0167] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the information distribution method provided by any one of the aforementioned method embodiments is implemented, including:

[0168] Receive feedback information from the object;

[0169] Determining reference classification information of each business system, and determining reference classification information that matches the feedback information as target classification information;

[0170] The feedback information is sent to the business system corresponding to the target classification information.

[0171] In some embodiments of the present application, the reference classification information includes: service description information;

[0172] The determining of reference classification information matching the feedback information as target classification information includes:

[0173] Determine the business description information corresponding to each business system, and obtain a set of business description information corresponding to all business systems;

[0174] Inputting the service description information in the service description information set and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the service description information set;

[0175] The target service description information is used as the target classification information.

[0176] In some embodiments of the present application, the reference classification information includes historical feedback information, where the historical feedback information is feedback information that has been historically processed by the business system;

[0177] The determining of reference classification information matching the feedback information as target classification information includes:

[0178] Determine the corresponding historical feedback information in each business system and obtain the historical feedback information set corresponding to all business systems;

[0179] Determining text similarity between the feedback information and any historical feedback information in the historical feedback information set;

[0180] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0181] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0182] If the number of historical feedback information whose text similarity meets the first predetermined condition is greater than a predetermined value, the business system corresponding to the historical feedback information whose text similarity meets the first predetermined condition is selected as a candidate business system;

[0183] Taking the service description information corresponding to the to-be-selected service system as the to-be-selected service description information, and obtaining a to-be-selected service description information set;

[0184] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0185] The target service description information is used as the target classification information.

[0186] In some embodiments of the present application, the reference classification information includes classification keywords, which are characters contained in feedback information that has been historically processed by the corresponding business system;

[0187] The determining of reference classification information matching the feedback information as target classification information includes:

[0188] Segmenting the feedback information to obtain a corresponding keyword set;

[0189] The classification keywords included in the keyword set are used as target classification information that matches the feedback information.

[0190] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0191] If the number of business systems corresponding to the classification keywords included in the keyword set is greater than a predetermined value, the business systems corresponding to the classification keywords included in the keyword set are used as candidate business systems;

[0192] Determine the corresponding historical feedback information in each candidate business system, and obtain a set of candidate historical feedback information corresponding to all candidate business systems;

[0193] Determining text similarity between the feedback information and any historical feedback information in the set of to-be-selected historical feedback information;

[0194] The historical feedback information whose text similarity meets the first predetermined condition is used as the target classification information.

[0195] In some embodiments of the present application, determining reference classification information that matches the feedback information as target classification information further includes:

[0196] If the number of business systems corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, determining the business description information corresponding to the candidate business system, and obtaining a set of candidate business description information corresponding to all candidate business systems;

[0197] Inputting the service description information in the set of to-be-selected service description information and the feedback information into a large language model, so that the large language model matches target service description information corresponding to the feedback information from the set of to-be-selected service description information;

[0198] The target service description information is used as the target classification information.

[0199] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0200] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, or of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment.

[0201] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "comprise", "include", "contain" and "have" are inclusive and therefore specify the presence of stated features, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the specific order described or illustrated, unless the order of execution is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0202] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is intended to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. An information distribution method, characterized in that: include: Receive feedback from the object; Determine reference classification information of each business system, and determine reference classification information matching the feedback information as target classification information, wherein the reference classification information is character information determined according to a functional description of a business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information; The feedback information is sent to the business system corresponding to the target classification information.

2. The method according to claim 1, characterized in that The reference classification information includes: business description information; The determining of reference classification information matching the feedback information as target classification information includes: Determine the business description information corresponding to each business system, and obtain a set of business description information corresponding to all business systems; Inputting the service description information in the service description information set and the feedback information into a large language model, so that the large language model matches the target service description information corresponding to the feedback information from the service description information set; The target service description information is used as the target classification information.

3. The method according to claim 1, characterized in that The reference classification information includes historical feedback information, and the historical feedback information is feedback information that has been historically processed by the business system; The determining of reference classification information matching the feedback information as target classification information includes: Determine the corresponding historical feedback information in each business system, and obtain the set of historical feedback information corresponding to all business systems; Determining text similarity between the feedback information and any historical feedback information in the historical feedback information set; The historical feedback information whose text similarity satisfies the first predetermined condition is used as the target classification information.

4. The method according to claim 3, characterized in that The step of determining the reference classification information matching the feedback information as the target classification information further includes: If the amount of the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, the business system corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is taken as a candidate business system; Taking the service description information corresponding to the service system to be selected as the service description information to be selected, and obtaining a set of service description information to be selected; Inputting the service description information in the service description information set to be selected and the feedback information into a large language model, so that the large language model matches the target service description information corresponding to the feedback information from the service description information set to be selected; The target service description information is used as the target classification information.

5. The method according to claim 1, characterized in that The reference classification information includes classification keywords, and the classification keywords are characters contained in the feedback information processed historically by the corresponding business system; The determining of reference classification information matching the feedback information as target classification information includes: Segmenting the feedback information to obtain a corresponding keyword set; The classification keywords included in the keyword set are used as target classification information that matches the feedback information.

6. The method according to claim 5, characterized in that The step of determining the reference classification information matching the feedback information as the target classification information further includes: If the number of business systems corresponding to the classification keywords included in the keyword set is greater than a predetermined value, the business systems corresponding to the classification keywords included in the keyword set are used as business systems to be selected; Determine the corresponding historical feedback information in each business system to be selected, and obtain a set of historical feedback information to be selected corresponding to all business systems to be selected; Determining text similarity between the feedback information and any historical feedback information in the set of historical feedback information to be selected; The historical feedback information whose text similarity satisfies the first predetermined condition is used as the target classification information.

7. The method according to claim 6, characterized in that The step of determining the reference classification information matching the feedback information as the target classification information further includes: If the number of business systems corresponding to the historical feedback information whose text similarity satisfies the first predetermined condition is greater than a predetermined value, then determining the business description information corresponding to the candidate business system, and obtaining a set of candidate business description information corresponding to all candidate business systems; Inputting the service description information in the service description information set to be selected and the feedback information into a large language model, so that the large language model matches the target service description information corresponding to the feedback information from the service description information set to be selected; The target service description information is used as the target classification information.

8. An information distribution system, characterized in that: include: An information receiving module, used for receiving feedback information from an object; An information classification module, used to determine reference classification information of each business system, and based on a matching strategy corresponding to the reference classification information, determine reference classification information matching the feedback information as target classification information, wherein the reference classification information is character information determined based on a functional description of a business of each business system and / or previously processed feedback information, and is used to classify the feedback information with reference to the character information; The information distribution module is used to send the feedback information to the business system corresponding to the target classification information.

9. A computer device: at least one communication interface; at least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; at least one memory connected to the at least one bus, wherein: The processor is configured to execute the information distribution method according to any one of claims 1 to 7.

10. A computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the information distribution method according to any one of claims 1 to 7.

Images