Method for handling customer complaints, related device and computer program product

Abstract

The present application provides a method, related device and computer program product for handling customer complaints. The application generates abbreviations of the customer complaint information in response to receiving customer complaint information sent by a user; in response to being unable to determine the first reply information corresponding to the abbreviation information based on the reply database, rewrites the entity words in the abbreviation information using a standard word library to generate a rewritten abbreviation information; pushes the rewritten abbreviation information to manual customer service; and in response to receiving a second reply information returned by the manual customer service for the rewritten abbreviation information, pushes the second reply information to the user. Thus, not only can the efficiency and quality of customer complaint handling be improved by combining and complementing the database and manual customer service answers, but also, in the process of providing a reply through the customer, it can help the manual customer service understand the content of the customer complaint more efficiently and with higher quality based on the rewriting of the customer complaint information, thereby improving the efficiency and quality of manual customer service's question answering.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and apparatus for handling customer complaints, an electronic device, a computer-readable medium, and a computer program product. Background Art

[0002] Customer complaints, or complaints from customers and users, typically refer to feedback, such as inquiries and comments, raised by users and clients regarding dissatisfaction with service or product quality during the service provider's service process. Alternatively, customers can use customer complaints to raise suggestions, objections, protests, claims, and requests for resolution of issues regarding services and products.

[0003] In this context, to better handle customer complaints, address their needs, and provide services to both users and customers, service providers often choose to collect customer complaint information through on-site and online communication. Subsequently, service providers can use this information to understand the problems and needs encountered by users. Based on these problems and needs, they can provide targeted advice and strategies to address these issues and enhance their experience with products and services. Therefore, how to more efficiently and effectively handle customer complaint information is a matter of concern and urgent need. Summary of the Invention

[0004] Multiple aspects of the present application provide a method, apparatus, electronic device, computer-readable storage medium, and computer program product for handling customer complaints, which can not only improve the efficiency and quality of customer complaint handling by combining and complementing a database and manual customer service responses, but also, in the process of providing responses through customers, can help manual customer service staff understand the content of customer complaints more efficiently and with higher quality based on rewriting customer complaint information, thereby improving the efficiency and quality of manual customer service staff's question answering.

[0005] One aspect of the present application provides a method for handling customer complaints, comprising: in response to receiving customer complaint information sent by a user, generating abbreviation information of the customer complaint information; in response to being unable to determine first reply information corresponding to the abbreviation information based on a reply database, rewriting entity words in the abbreviation information using a standard word library to generate rewritten abbreviation information; pushing the rewritten abbreviation information to manual customer service; and in response to receiving second reply information returned by the manual customer service for the rewritten abbreviation information, pushing the second reply information to the user.

[0006] Another aspect of the present application provides an apparatus for handling customer complaints, comprising: an abbreviation information generation unit, configured to generate abbreviation information of the customer complaint information in response to receiving customer complaint information sent by a user; a rewriting information generation unit, configured to rewrite entity words in the abbreviation information using a standard word library to generate rewritten abbreviation information in response to an inability to determine first reply information corresponding to the abbreviation information based on a reply database; a rewriting information push unit, configured to push the rewritten abbreviation information to manual customer service; and a first reply push unit, configured to push the second reply information to the user in response to receiving second reply information returned by the manual customer service for the rewritten abbreviation information.

[0007] Another aspect of the present application provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable 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 execute the method for handling customer complaints provided above.

[0008] In another aspect of the present application, a computer-readable storage medium is provided, on which computer program instructions are stored. The computer program instructions can be executed by a processor to implement the method for handling customer complaints provided above.

[0009] In another aspect of the present application, a computer program product includes a computer program having computer program instructions stored thereon, and when the computer program is executed by a processor, the computer program can implement the method for handling customer complaints provided above.

[0010] In the solution provided by the embodiment of the present application, in response to receiving customer complaint information sent by a user, a summary of the customer complaint information is generated; in response to being unable to determine the first reply information corresponding to the summary information based on the reply database, the entity words in the summary information are rewritten using the standard word library to generate the rewritten summary information; the rewritten summary information is pushed to the manual customer service; in response to receiving the second reply information returned by the manual customer service for the rewritten summary information, the second reply information is pushed to the user. In this way, not only can the efficiency and quality of customer complaint handling be improved by combining and complementing the database and the manual customer service answers, but also, in the process of providing a reply through the customer, it can also help the manual customer service to understand the content of the customer complaint more efficiently and with higher quality based on the rewriting of the customer complaint information, thereby improving the efficiency and quality of the manual customer service's question answering. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, a brief introduction will be given below to the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0012] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0013] Figure 1 A schematic diagram of a process for handling customer complaints provided in one embodiment of the present application;

[0014] Figure 2 A schematic diagram of a process for generating abbreviations of customer complaint information provided in one embodiment of the present application;

[0015] Figure 3 A schematic diagram of an exemplary architecture for implementing a method for handling customer complaints in an application scenario provided in one embodiment of the present application;

[0016] Figure 4 A schematic diagram of the structure of an apparatus for handling customer complaints provided in one embodiment of the present application;

[0017] Figure 5 The figure is a schematic diagram of the structure of an electronic device suitable for implementing the solution in the embodiment of the present application.

[0018] The same or similar reference numerals in the drawings represent the same or similar components. DETAILED DESCRIPTION

[0019] 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.

[0020] In a typical configuration of the present application, the terminal and the equipment of the service network each include one or more processors (CPUs), input / output interfaces, network interfaces and memories.

[0021] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0022] Computer-readable media include both permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology for information storage. Information can be computer program instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc-read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0023] As discussed above, how to handle customer complaints more efficiently and with higher quality is worthy of attention and an urgent need.

[0024] In some solutions, service providers usually choose to provide handling strategy manuals and help manuals based on the content of customer complaints, so that users can find solutions by querying the strategy manuals and help manuals when they encounter customer complaints. For example, a query manual can be constructed in a "question-answer" question-answer pair format, so that when users encounter customer complaints, they can match the specific content of the incident with the "questions" recorded in the query manual. Then, through the questions hit by the matching results, the target question-answer pairs corresponding to the questions in the query manual are used to find the answers. Then, the solution strategy is determined based on the target answers. Alternatively, the service provider will configure manual customer service for potential customer complaints, so that users can choose to obtain corresponding handling strategies through manual interaction.

[0025] However, these approaches, such as those based on manual queries, often require users to accurately identify the problem, which places excessive demands on users and is time-consuming. Manually providing interactive solutions also places high demands on the communication and question-answering skills of human customer service representatives. Otherwise, it can be difficult for users to obtain solutions efficiently and effectively.

[0026] In this regard, an embodiment of the present application provides a method for handling customer complaints, which generates abbreviations of the customer complaint information in response to receiving customer complaint information sent by a user; in response to being unable to determine the first reply information corresponding to the abbreviation information based on the reply database, rewrites the entity words in the abbreviation information using a standard word library to generate a rewritten abbreviation information; pushes the rewritten abbreviation information to manual customer service; and in response to receiving the second reply information returned by the manual customer service for the rewritten abbreviation information, pushes the second reply information to the user. Thus, not only can the efficiency and quality of customer complaint handling be improved by combining and complementing the database and manual customer service answers, but also, in the process of providing a reply through the customer, it can also help manual customer service understand the content of the customer complaint more efficiently and with higher quality based on the rewriting of the customer complaint information, thereby improving the efficiency and quality of manual customer service's question answering.

[0027] In practical scenarios, the method can be executed by a user device, a device formed by integrating a user device and a network device via a network, or an application running on such a device. User devices include, but are not limited to, computers, mobile phones, tablets, smart watches, wristbands, and other terminal devices. Network devices include, but are not limited to, network hosts, single network servers, multiple network server clusters, or cloud computing-based computer collections. Here, the cloud consists of a large number of hosts or network servers based on cloud computing. Cloud computing is a type of distributed computing, consisting of a group of loosely coupled computers forming a virtual computer.

[0028] When the execution subject is software, it can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules, or as a single software or software module, and is not specifically limited here.

[0029] Figure 1 A process 100 for handling customer complaints provided by an embodiment of the present application is shown. The process 100 includes at least the following processing steps:

[0030] (Step) S101 : In response to receiving customer complaint information sent by a user, generating a summary of the customer complaint information.

[0031] In an embodiment of the present application, the execution entity responds to customer complaint information sent by a user and, based on processing of the customer complaint information, obtains a summary of the customer complaint information. For example, after obtaining the customer complaint information, the execution entity may perform word segmentation and segmentation operations on the customer complaint information. Then, from the segmentation results, characters, phrases, or sentences that do not affect the "semantic content" of the customer complaint information are deleted to achieve a reduction purpose and obtain a summary.

[0032] In some embodiments, the execution entity may also choose to extract keywords from the customer complaint information and then rewrite the combination of keywords to obtain abbreviated information (for example, determining the subject, predicate, and object of the abbreviated information based on the keywords to obtain the abbreviated information).

[0033] In some embodiments, the execution entity may also choose to generate a summary of the customer complaint information based on the semantic analysis results of the customer complaint information.

[0034] The abbreviated information includes the semantics of key customer complaint information (e.g., the product being complained about, the problem it encountered). The abbreviated information is typically shorter than the actual customer complaint information. This allows the abbreviated information to convey the full semantics of the customer complaint information while also shortening its length. In some embodiments, the implementing entity may also choose to delete or block meaningless words in the customer complaint information, such as interjections and greetings, to achieve the goal of reducing the customer complaint information and determining the abbreviated version.

[0035] Therefore, the executing entity can use the abbreviated information to more accurately understand the content of the customer complaint information. In other words, the abbreviated information can also be understood and regarded as an "expression optimization" of the customer complaint information, so that the subsequent executing entity or manual customer service can more clearly and accurately determine the content of the customer complaint.

[0036] In some embodiments, the customer complaint information sent by users may be directly submitted or uploaded by the users, allowing the execution entity to communicate the user's customer complaint information directly through communication with the user. In some embodiments, to reduce the frequency of data exchange and alleviate the access pressure on the execution entity, the execution entity may also choose to utilize a cache to collect the customer complaint information sent by users, thereby alleviating the communication pressure on the execution entity. For example, multiple caches may be pre-maintained (e.g., constructed using random access memory (ROM)). These caches then store customer complaint information sent by (different) users. In other words, users can provide customer complaint information by communicating with the cache. The execution entity can then request and obtain customer complaint information from the (target) cache by communicating with it. This reduces the number of communications between the user and the execution entity, reducing the communication pressure on the execution entity.

[0037] In some embodiments, based on the above considerations, the cache library can also be optimized for communication capabilities so that it can communicate with users more efficiently and stably, improving the user's customer complaint information upload and feedback experience.

[0038] Accordingly, in some embodiments, the executing entity may request customer complaint information sent by users from the cache based on a preset retrieval cycle time condition. For example, upon detecting that the current moment satisfies the preset retrieval cycle time condition (for example, the retrieval cycle time condition may correspond to a condition configured for the target cache, such as every one day. Accordingly, if the executing entity determines that the interval between the current moment and the last retrieval is one day, it determines that the preset retrieval cycle time condition is detected to be met), the executing entity may send a retrieval request to the target cache to request the customer complaint information sent by users. Accordingly, the target cache may respond to the retrieval request and then return the customer complaint information sent by users that it has locally stored to the executing entity.

[0039] It should be understood that the target cache library can also judge locally whether the acquisition cycle time condition is met. Accordingly, the target cache library can choose to provide the stored customer complaint information sent by the user to the execution entity only when the target cache library also determines locally that the acquisition cycle time condition is met.

[0040] Compared to the above-mentioned "temporary communication" method (i.e., communicating and obtaining customer complaint information sent by users only when time period conditions are met), in some embodiments, the execution entity can also choose to establish continuous communication with a target cache library used to store at least one customer complaint information sent by users.

[0041] Then, the execution entity can monitor the storage status of the target cache library based on continuous communication. For example, the execution entity can monitor the usage of the storage in the target cache library, that is, the number of customer complaint information sent by the stored users. If the execution entity monitors that the storage status indicates that the number of customer complaint information sent by users currently stored in the target cache library exceeds a preset number threshold (for example, the number threshold can be determined based on the storage capacity and storage space upper limit of the target cache library, for example, the number threshold can be determined in advance based on the "number of storage items" corresponding to the risk of storage removal in the target cache library), the execution entity can respond to this and choose to communicate with the target cache library and request the target cache library to obtain at least one customer complaint information sent by the user.

[0042] Therefore, the execution entity can, based on the storage status of the target cache library, actively and promptly request the target cache library to currently store the customer complaint information sent by the user when there may be a risk of data overflow, thereby ensuring data security (for example, avoiding the loss of user customer complaint information due to data overflow in the target cache library).

[0043] In some embodiments, after the target cache library returns and provides the customer complaint information sent by the user to the execution entity, the target cache library may also choose to clear the customer complaint information sent by the user locally to free up the local storage space of the target cache library and improve efficiency.

[0044] Alternatively or additionally, to prevent data from being mistakenly cleared, the target cache library may, after providing the executing entity with the customer complaint information sent by the user, wait for the executing entity to return a confirmation instruction, and execute the clearing operation upon receipt of the confirmation instruction. For example, upon determining that the customer complaint information has been transferred to the executing entity's local cache, or upon completing processing of the customer complaint information and transferring the customer complaint information and corresponding processing results to a local / non-local storage device for storing the processed customer complaint information, the executing entity may choose to communicate the above-mentioned confirmation instruction to the target cache library, thereby allowing the target cache library to clear the corresponding customer complaint information sent by the user.

[0045] S102: Determine whether the first reply information corresponding to the abbreviated information can be determined based on the reply database.

[0046] In an embodiment of the present application, after obtaining the abbreviated information based on the above S101, the executing entity determines whether it can determine the reply information corresponding to the abbreviated information based on the reply database (for ease of understanding, it can be described as the first reply information). The reply database can be constructed in advance based on an expert knowledge graph or other methods, and the reply database stores solution strategies corresponding to customer complaint information (or customer complaint issues) (for example, for equipment failures, the solution strategy can be a troubleshooting solution provided based on the failure type of the equipment failure). For example, in the reply database, customer complaint issues and corresponding solution strategies can be maintained in the form of "reference questions-reference answers". Generally, such an expert knowledge graph can be constructed based on at least the combined information of events and processing solutions provided by the service provider.

[0047] In some embodiments, for the expert knowledge graph constructed based on the combined information of events and processing solutions provided by the service provider, the executing entity can also enrich the corresponding "processing solutions" that can be selected based on the search results of the processing solutions for the "events" (for example, the data results of searching and crawling in the open source knowledge base) to improve the content quality of the expert knowledge graph and the response database.

[0048] In an embodiment of the present application, the reply database may also store historical reply results of manual customer service and historical customer complaint information corresponding to the historical reply results. This allows the content in the reply database to be continuously enriched by reusing and referring to the reply results of manual customer service during historical interactions.

[0049] In some embodiments, for some customer complaint information that can be determined to be not within the scope of services provided by the service provider (for example, the event actually pointed to by the customer complaint information, or the cause of the event, is not within the scope of services provided by the service provider), a response solution strategy can also be configured (for example, providing pre-configured reply scripts). In this way, these customer complaint information that is not within the scope of services provided by the service provider can be "filtered" and "screened out" to reduce the labor cost of subsequent manual customer service.

[0050] Accordingly, the execution entity can use the abbreviated information as a "customer complaint question" and search and match it in the response database to determine whether a strategy corresponding to the "customer complaint question" can be output. For example, the execution entity can query whether a "reference question" that meets the required similarity with the "customer complaint question" can be found in the response database. If such a "reference question" is found, the execution entity can determine that the abbreviated information is matched in the response database and use the reference answer corresponding to the matched reference question as the processing result corresponding to the abbreviated information. In other words, the execution entity uses the abbreviated information as the first response information mentioned above.

[0051] In this step, if the execution subject cannot use the reply database to match the first reply information corresponding to the abbreviated information and cannot determine it, it can choose to execute S103.

[0052] S103: rewrite the entity words in the abbreviation information using a standard word library to generate rewritten abbreviation information.

[0053] In an embodiment of the present application, this step can be executed when the executing entity determines based on S102 that the first reply information corresponding to the abbreviation information cannot be matched using the reply database. In this step, the executing entity can use the standard vocabulary library to rewrite the entity words in the abbreviation information to generate the rewritten abbreviation information.

[0054] The standard word library records "standard entity words" used to normalize entities that express the same meaning but have different textual forms. For example, for entity words A1 and A2 with the same meaning, the execution entity can unify both into standard entity word A. In this way, the abbreviation information is rewritten using standard entity words.

[0055] In some embodiments, standard entity words can be pre-determined based on interactions with human customer service representatives and statistical results of "standard entity words." For example, "standard entity words" can be pre-determined using a reference vocabulary set provided by human customer service representatives, a human customer service working dictionary, and so on. Thus, by rewriting the abbreviation information, it can be expressed as "standard entity words" as much as possible.

[0056] S104: Push the rewritten abbreviated information to manual customer service.

[0057] In an embodiment of the present application, after obtaining the rewritten abbreviated information based on the above S103, the execution entity can push the rewritten abbreviated information to the human customer service representative (for example, by pushing the rewritten abbreviated information to the target device used by the human customer service representative based on a pre-configured communication path). Thus, the execution entity can rewrite the abbreviated information based on, for example, entity word normalization, so that the rewritten abbreviated information can at least be read more accurately and efficiently by the human customer service representative. This facilitates the human customer service representative to more accurately and efficiently understand and process customer complaint information.

[0058] In some embodiments, after the execution entity pushes the rewritten abbreviated information to the human customer service representative, it may also choose to send a progress synchronization message to the user. This progress synchronization message indicates that the customer complaint information has been pushed to the human customer service representative for processing. This allows users to understand the processing status and progress of their customer complaint information in real time, making it easier for users to resolve their customer complaint issues (for example, based on this progress synchronization message, users can determine that the customer complaint information has been transferred to the human customer service representative for processing and proactively choose to communicate with the human customer service representative to resolve the customer complaint issue).

[0059] S105 , in response to receiving the second reply information returned by the manual customer service in response to the rewritten thumbnail information, push the second reply information to the user.

[0060] In an embodiment of the present application, if the manual customer service returns a reply message (described as a second reply message for ease of understanding) in response to the rewritten abbreviated message, the execution entity can respond by pushing the second reply message to the user to complete the processing and reply to the user's customer complaint.

[0061] Accordingly, in some embodiments, in the above step S102, if the executing entity determines that the first reply information corresponding to the thumbnail information can be determined based on the reply database, the executing entity may directly choose to push the first reply information to the user.

[0062] For example, in Figure 1 The process shown may further include S106 , pushing the first reply information to the user.

[0063] S106 can be executed when the execution subject determines based on S102 that the first reply information corresponding to the abbreviated information can be determined based on the reply database, so as to use the first reply information as a reply to the customer complaint information and complete the customer complaint handling for the user.

[0064] Therefore, by combining the "machine reply" results based on the reply database with the manual reply results of the manual customer service, the efficiency of processing customer complaint information and the ability to process customer complaint information can be improved while ensuring the quality of customer complaint information processing. Accordingly, in the embodiments of the present application, the execution entity can also optimize the "customer complaint information" by "obtaining abbreviated information." In other words, by using abbreviated information with more "refined" content quality, the execution entity can more accurately perform matching in the reply database, avoiding the impact of matching accuracy and mismatching caused by the "lengthy" content in the customer complaint information provided by the user.

[0065] Then, the method for handling customer complaints provided by the present application generates abbreviations of the customer complaint information in response to receiving the customer complaint information sent by the user; in response to being unable to determine the first reply information corresponding to the abbreviation information based on the reply database, rewrites the entity words in the abbreviation information using the standard word library to generate a rewritten abbreviation information; pushes the rewritten abbreviation information to the manual customer service; and in response to receiving the second reply information returned by the manual customer service for the rewritten abbreviation information, pushes the second reply information to the user. In this way, not only can the efficiency and quality of customer complaint handling be improved by combining and complementing the database and the manual customer service answers, but also, in the process of the customer providing a reply, it can also help the manual customer service to understand the content of the customer complaint more efficiently and with higher quality based on the rewriting of the customer complaint information, thereby improving the efficiency and quality of the manual customer service's question answering.

[0066] In some embodiments, during the actual process of determining whether to handle the request by human customer service, that is, during the process of pushing the rewritten abbreviated information to the human customer service, the executing entity may also use the rewritten abbreviated information for "reconfirmation." That is, after obtaining the rewritten abbreviated information, the executing entity may further determine whether a reply corresponding to the rewritten abbreviated information can be determined based on the reply database (for ease of understanding, this will be described as third reply information). For example, after obtaining the rewritten abbreviated information, the executing entity may also determine whether the "rewritten abbreviated information" can be used to match a reference answer from the reply database. If so, the executing entity may still provide the "answer" matched using the reply database as a reply (i.e., third reply information) to the user.

[0067] If not, the executing entity can, as discussed above, choose to push the rewritten abbreviated information to a human customer service representative, who can then resolve the customer complaint and provide a response. This prevents "missed matches" caused by, for example, inaccurate recognition of the abbreviated information, which could lead to customer complaint information that could actually be responded to based on the response database being mistakenly pushed to a human customer service representative, thereby wasting time on the human customer service representative.

[0068] In some embodiments, as discussed above, the content of customer complaint information sent by users may vary. In particular, some customer complaint information may not actually be intended to resolve the corresponding "problem," meaning that such "customer complaint information" may be worthless. For example, users may simply use "customer complaint information" to vent their emotions, or use spam to attack or disrupt the execution entity. Therefore, upon receiving a customer complaint information sent by a user, the execution entity can pre-analyze and determine whether to generate a summary based on the information for subsequent processing based on the information.

[0069] For this, please refer to Figure 2 . Figure 2 A process 200 for generating a summary of customer complaint information provided by an embodiment of the present application is shown. The process 200 includes at least the following processing steps:

[0070] S201: In response to receiving customer complaint information sent by a user, detecting the number of preset type words included in the customer complaint information;

[0071] Specifically, when the execution entity receives the customer complaint information sent by the user, it can read the text content of the customer complaint information, and then detect the number of preset type words included in the customer complaint information. The preset type words can be determined based on a pre-configured screening strategy. For example, the preset type words may be "unfriendly" type words, such as "negative words" that do not meet the community user standards. Alternatively, the preset type words may be "words" with no practical meaning or unrecognizable meanings based on different scenarios. For example, the words may be "words" that are not commonly used in society (for example, certain traditional Chinese characters, words or symbols whose circulation rate or usage rate in society is lower than the circulation rate threshold or usage rate threshold, etc.). Accordingly, the service provider can determine the actual content corresponding to the "preset type words" based on screening requirements and screening strategies.

[0072] Furthermore, the execution entity may detect whether the customer complaint information includes such preset type words, and the number of such preset type words included.

[0073] If the executing entity determines that the number of preset type words included in the customer complaint information is less than or equal to a preset quantity threshold (for example, the quantity threshold can be configured based on the service provider's "tolerance level" for the preset type words. For example, the service provider may be able to tolerate the preset type words appearing X times in the customer complaint information, and "X" can be configured as the preset quantity threshold), it can choose to execute S202.

[0074] S202: Generate abbreviated information of the customer complaint information.

[0075] This step is similar to the process of generating thumbnail information in the above S101. For related content, reference may be made to the content discussed in the above S101, and will not be repeated here.

[0076] Alternatively or additionally, in some embodiments, process 200 may further include S203, which is executed when the execution subject determines that the number of words of a preset type is greater than a preset number threshold.

[0077] S203, refusing to generate abbreviated information;

[0078] Specifically, "customer complaint information" containing more than a preset threshold of predefined terms can be considered "pseudo-customer complaint information" because it may not conform to regulations and lacks customer intent. The execution entity can refuse to process such "pseudo-customer complaint information" as abbreviated information to reduce potential waste of computing resources. Accordingly, the execution entity can continue with S204.

[0079] S204: Send prompt information regarding the customer complaint information to the user.

[0080] Specifically, the reason the implementing entity refuses to generate abbreviated information can actually be understood as the customer complaint information lacking "substantial content" or containing too many non-compliant words. Therefore, if the abbreviated information is refused, the implementing entity can communicate with the user to send a prompt message to inform them of the "non-compliance" or "lack of substance" situation, thereby assisting the user in adjusting the "customer complaint information" to better and more accurately express the customer's complaint needs.

[0081] In some embodiments, if the executing entity refuses to process the customer complaint information as abbreviated information, the executing entity can also synchronize this situation to the user (in some scenarios, the reason for the rejection can also be provided to the user) to more accurately prompt the reason for the rejection and facilitate the user to make adjustments.

[0082] Based on any of the above embodiments, to improve performance, the executing entity may choose to utilize a generative language model to process at least one of the following processes: generating a summary of the customer complaint information, querying the first and third responses from the response database, and rewriting the summary to obtain the rewritten summary. For example, the summary and / or the rewritten summary may be generated by a generative language model.

[0083] A general language model (GLM) is a type of large language model (LLM), also known as a large language model. It is an artificial intelligence model designed to understand and generate human language. Furthermore, a GLM can perform processing operations based on its understanding to produce corresponding results. While other types of LLMs can be used to equivalently implement the processing steps implemented by a GLM, this discussion focuses solely on the GLM.

[0084] LLMs are characterized by their large size and can typically include a large number of parameters to help them learn complex patterns in language data. These models are often based on deep learning architectures such as transformers, which helps them provide better processing performance on various NLP tasks.

[0085] As discussed above, in an embodiment of the present application, the execution entity may also choose to utilize GLM to improve processing performance. For example, in the process of generating abbreviated information of customer complaint information, GLM may be used to generate abbreviated information of customer complaint information. For example, the generative language model may be pre-trained so that it has the ability to generate abbreviated information of customer complaint information. In this case, the execution entity may instruct the generation of abbreviated information of customer complaint information based on pre-configured guide words, guide tags, etc. For example, in generating abbreviated information of customer complaint information, the guide words may be, for example: "keep the semantics of the customer complaint information unchanged and shorten the length of the text content of the customer complaint information", "extract the core content of the customer complaint information and generate abbreviated information", etc.

[0086] Similarly, for the generative language model, it can be configured by default to omit the "guide word". For example, for the purpose of rewriting the answer semantics into standard answer information of standard length using a preset knowledge base, the generative language model can, based on the default configuration, naturally understand the need to generate abbreviations of customer complaint information. Thus, by using the default configuration, the generative language model can stably and directionally generate abbreviations of customer complaint information. Similarly, the execution subject can also use different guide words to use the generative language model to implement at least one of the processes of generating abbreviations of customer complaint information, querying the first reply information and the third reply information using the reply database, and rewriting the abbreviation information to obtain the rewritten abbreviation information.

[0087] Therefore, the generative language model can be used to more efficiently and efficiently realize at least one of the following processes: generating abbreviation information of customer complaint information, querying the first reply information and the third reply information using the reply database, and rewriting the abbreviation information to obtain the rewritten abbreviation information.

[0088] To deepen understanding, this disclosure also provides an exemplary architecture for the specific application of this application in conjunction with a specific application scenario. Figure 3 , Figure 3 An exemplary architecture 300 for implementing a method for handling customer complaints in an application scenario provided by an embodiment of the present application is shown.

[0089] In the architecture 300 , the server 310 can be used as an executor of the method for handling customer complaints, for example.

[0090] Typically, server 310 can at least communicate with a user's terminal device 320 to obtain customer complaint information 325 sent by the user. However, for ease of understanding, in architecture 300, the execution entity can also choose to utilize cache 330 to collect "customer complaint information 325." For example, a user can provide customer complaint information 325 to cache 330 for storage based on communication between terminal device 320 and cache 330. Subsequently, server 310 can request and obtain the customer complaint information 325 based on communication with cache 330.

[0091] Then, after the server 310 obtains the customer complaint information 325 , the LLM 315 may be used to process the customer complaint information 325 to generate a summary of the customer complaint information 340 .

[0092] Server 310 may then use LLM 315 to instruct it to determine whether a response database (not shown) can be used to determine an answer (e.g., first response information 350) corresponding to abbreviated information 340. If LLM 315 determines that first response information 350 can be determined using the response database, server 310 may provide first response information 350 to terminal device 320 as a "reply" to customer complaint information 325 sent by the user.

[0093] Accordingly, if LLM 315 determines that the first reply information 350 cannot be determined using the "reply database", the server 310 can instruct LLM 315 to rewrite the entity words in the abbreviation information 340 using the "standard word library" (not shown in the figure) to generate the rewritten abbreviation information 345.

[0094] Then, the server 310 communicates the rewritten abbreviated information 345 to the human customer service 360 ​​.

[0095] Accordingly, if the server 310 receives a second reply message 355 returned by the manual customer service 360 ​​in response to the rewritten abbreviated information 345 , the server 310 may provide the second reply message 355 to the terminal device 320 as a “reply” to the customer complaint message 325 sent by the user.

[0096] The present application also provides a device for handling customer complaints, the structure of which is as follows: Figure 4 The device 400 shown. The device 400 includes: an abbreviation information generating unit 410, configured to generate abbreviation information of the customer complaint information in response to receiving customer complaint information sent by a user; a rewriting information generating unit 420, configured to rewrite the entity words in the abbreviation information using a standard word library to generate rewritten abbreviation information in response to failure to determine the first reply information corresponding to the abbreviation information based on the reply database; a rewriting information pushing unit 430, configured to push the rewritten abbreviation information to manual customer service; and a first reply pushing unit 440, configured to push the second reply information to the user in response to receiving the second reply information returned by the manual customer service for the rewritten abbreviation information.

[0097] In some embodiments, the rewritten information pushing unit 430 is further configured to, in response to failure to determine third reply information corresponding to the rewritten abbreviated information based on the reply database, push the rewritten abbreviated information to manual customer service.

[0098] In some embodiments, the apparatus 400 further includes: a second reply pushing unit configured to push the first reply information to the user in response to being able to determine the first reply information corresponding to the thumbnail information based on the reply database.

[0099] In some embodiments, the abbreviation information generation unit 410 includes: a word detection subunit, configured to detect the number of preset type words included in the customer complaint information in response to receiving customer complaint information sent by the user; and an abbreviation information generation subunit, configured to generate abbreviation information of the customer complaint information in response to the number being less than or equal to a preset number threshold.

[0100] In some embodiments, the device 400 further includes: a prompt information sending unit configured to refuse to generate abbreviated information in response to the number being greater than a preset number threshold; and send a prompt information regarding the customer complaint information to the user.

[0101] In some embodiments, the device 400 also includes: a first customer complaint information request unit, which is configured to send an acquisition request for requesting customer complaint information sent by the user to the target cache library in response to the current moment satisfying a preset acquisition cycle time condition, wherein the customer complaint information sent by the user is returned to the local by the target cache library in response to the acquisition request.

[0102] In some embodiments, the device 400 also includes: a communication establishing unit, configured to establish continuous communication with a target cache library, wherein the target cache library is used to store at least one customer complaint information sent by a user; a storage status monitoring unit, configured to monitor the storage status of the target cache library based on continuous communication; a second customer complaint information requesting unit, configured to request the target cache library to obtain at least one customer complaint information sent by a user in response to a storage status indication that the number of customer complaint information sent by users currently stored in the target cache library exceeds a preset number threshold.

[0103] In some embodiments, the device 400 also includes: a progress synchronization unit, which is configured to send progress synchronization information to the user after pushing the rewritten thumbnail information to the manual customer service, wherein the progress synchronization information is used to indicate that the customer complaint information has been pushed to the manual customer service for processing.

[0104] In some embodiments, the abbreviated information and / or the rewritten abbreviated information is generated by a generative language model.

[0105] Based on the same inventive concept, an electronic device, a readable storage medium, and a computer program product are also provided in an embodiment of the present application. The method corresponding to the electronic device can be the method for handling customer complaints in the aforementioned embodiment, and its principle of solving the problem is similar to that of the aforementioned embodiment. The electronic device provided in an embodiment of the present application includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable 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 execute the methods and / or technical solutions of the aforementioned multiple embodiments of the present application.

[0106] An electronic device can be a user device, or a device formed by integrating a user device and a network device via a network, or an application running on any of the above devices. User devices include, but are not limited to, computers, mobile phones, tablets, smart watches, wristbands, and other terminal devices. Network devices include, but are not limited to, network hosts, single network servers, multiple network servers, or a collection of computers based on cloud computing, and can be used to implement some of the processing functions required for setting an alarm. Here, the cloud is composed of a large number of hosts or network servers based on cloud computing. Cloud computing is a type of distributed computing, consisting of a group of loosely coupled computers forming a virtual computer.

[0107] Figure 5The structure of an electronic device suitable for implementing the methods and / or technical solutions in the embodiments of the present application is shown. The electronic device 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 502 or programs loaded from a storage unit 508 into a random access memory (RAM) 503. RAM 503 also stores various programs and data required for system operation. CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to bus 504.

[0108] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, a touch screen, a microphone, an infrared sensor, and the like; an output section 507 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), an LED display, an OLED display, and speakers; a storage section 508 including one or more computer-readable media such as a hard disk, an optical disk, a magnetic disk, and a semiconductor memory; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 509 performs communication processing via a network such as the Internet.

[0109] In particular, the methods and / or embodiments of the present application can be implemented as computer software programs. For example, the embodiments disclosed herein include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the method illustrated in the flowchart. When the computer program is executed by the central processing unit (CPU) 501, the aforementioned functions defined in the method of the present application are performed.

[0110] Another embodiment of the present application further provides a computer-readable storage medium and a computer program product, on which computer program instructions are stored. The computer program instructions can be executed by a processor to implement the methods and / or technical solutions of any one or more embodiments of the present application.

[0111] Specifically, this embodiment may employ any combination of one or more computer-readable media. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may, for example, be a system, device, or component including, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, 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 thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0112] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0113] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0114] Computer program code for performing the operations of the present application can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0115] The flow chart or block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the equipment, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code include one or more executable instructions for realizing the logical function of the specification. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated system for hardware that performs the function or operation of the specification, or can be implemented with a combination of dedicated hardware and computer instructions.

[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules and units is only a logical function division. There may be other division methods in actual implementation. For example, with units as an example, for example, multiple units or page components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0118] Units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0119] In addition, the functional modules and units in the various embodiments of the present application may be integrated into a single processing module or unit, or each module or unit may exist physically separately, or two or more units may be integrated into a single module or unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional modules or units.

[0120] The above-mentioned integrated modules and units implemented in the form of software functional modules and units can be stored in a computer-readable storage medium. The above-mentioned software functional modules and units are stored in a storage medium and include a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to perform some steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, optical disks, and other media that can store program code.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0122] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device through software or hardware. Terms such as "first" and "second" are used to indicate names and do not imply any particular order.

Claims

1. A method for handling customer complaints, comprising: In response to receiving customer complaint information sent by a user, generating a summary of the customer complaint information; In response to a failure to determine the first reply information corresponding to the abbreviated information based on the reply database, rewriting the entity words in the abbreviated information using a standard word library to generate rewritten abbreviated information; Pushing the rewritten abbreviated information to manual customer service; In response to receiving the second reply information returned by the manual customer service in response to the rewritten abbreviated information, the second reply information is pushed to the user.

2. The method according to claim 1, wherein The step of pushing the rewritten abbreviated information to manual customer service includes: In response to failure to determine third reply information corresponding to the rewritten abbreviated information based on the reply database, the rewritten abbreviated information is pushed to manual customer service.

3. The method according to claim 1, further comprising: In response to being able to determine first reply information corresponding to the abbreviated information based on the reply database, the first reply information is pushed to the user.

4. The method according to claim 1, wherein The step of generating a summary of the customer complaint information in response to receiving the customer complaint information sent by the user includes: In response to receiving customer complaint information sent by a user, detecting the number of words of a preset type included in the customer complaint information; In response to the number being less than or equal to a preset number threshold, abbreviated information of the customer complaint information is generated.

5. The method according to claim 4, further comprising: In response to the number being greater than the preset number threshold, refusing to generate the abbreviated information; Sending a prompt message regarding the customer complaint information to the user.

6. The method according to claim 1, further comprising: In response to the current moment satisfying the preset acquisition cycle time condition, an acquisition request for requesting the customer complaint information sent by the user is sent to the target cache library, wherein the customer complaint information sent by the user is returned to the local by the target cache library in response to the acquisition request.

7. The method according to claim 1, further comprising: Establishing continuous communication with a target cache, wherein the target cache is used to store at least one customer complaint message sent by the user; Monitoring a storage status of the target cache based on the continuous communication; In response to the storage status indicating that the number of customer complaint information sent by the user currently stored in the target cache exceeds a preset number threshold, a request is made to the target cache to obtain at least one customer complaint information sent by the user.

8. The method according to claim 1, after pushing the rewritten abbreviated information to manual customer service, further comprising: Send progress synchronization information to the user, wherein the progress synchronization information is used to indicate that the customer complaint information has been pushed to the manual customer service for processing.

9. The method according to any one of claims 1 to 8, wherein The abbreviation information and / or the rewritten abbreviation information is generated by a generative language model.

10. A device for handling customer complaints, comprising: an abbreviation information generating unit, configured to generate abbreviation information of the customer complaint information in response to receiving the customer complaint information sent by the user; The rewriting information generating unit is configured to, in response to failure to determine the first reply information corresponding to the abbreviated information based on the reply database, rewrite the entity words in the abbreviated information using a standard word library to generate rewritten abbreviated information: A rewritten information pushing unit is configured to push the rewritten abbreviated information to a manual customer service representative; The first reply pushing unit is configured to push the second reply information to the user in response to receiving the second reply information returned by the manual customer service in response to the rewritten abbreviated information.

11. An electronic device, comprising: 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 method according to any one of claims 1 to 9.

12. A computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions can be executed by a processor to implement the method according to any one of claims 1 to 9.

13. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images