Knowledge base updating method and device, storage medium and electronic equipment

Abstract

This application discloses a method, apparatus, storage medium, and electronic device for updating a knowledge base. It relates to the field of financial technology or other related fields. The method includes: acquiring audio recordings within a preset time period and corresponding agent operation logs; obtaining question keywords based on the audio recordings; querying agent operation records for the knowledge base from the agent operation logs based on the question keywords, wherein the operation records at least include a timestamp of the agent's response to the question; determining the response content and target customer feedback based on the timestamps and audio recordings; and updating the knowledge base based on the response content and feedback information to obtain an updated knowledge base. This application solves the technical problem of low update efficiency of agent knowledge bases due to manual updates in related technologies.

Description

Technical Field

[0001] This application relates to the field of financial technology or other related fields, and more specifically, to a method, apparatus, storage medium, and electronic device for updating a knowledge base. Background Technology

[0002] Currently, financial institutions primarily update their agent knowledge bases manually. For example, staff manually sample and analyze call recordings to identify customer hot topics and agent responses, then manually update the knowledge base accordingly. However, this manual updating method is not only labor-intensive but also time-consuming, resulting in low efficiency in both labor and time costs.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and electronic device for updating a knowledge base, to at least solve the technical problem in the related art where updating the agent knowledge base manually results in low updating efficiency.

[0005] According to one aspect of this application, a method for updating a knowledge base is provided, comprising: acquiring audio recordings within a preset time period and corresponding agent operation logs, wherein the audio recordings are voice call information between a financial institution's customer service representative and a target customer, and the agent operation logs are operation logs of the customer service representative querying the answer content of a question from the knowledge base; acquiring question keywords based on the audio recordings, wherein the question keywords are keywords in a question raised by the target customer; querying the agent's operation records on the knowledge base from the agent operation logs based on the question keywords, wherein the operation records include at least a timestamp of the agent's query of the answer content of the question; determining the answer content of the question and the target customer's feedback information based on the timestamp and the audio recordings, wherein the feedback information is used to characterize the target customer's satisfaction with the answer content; and updating the knowledge base based on the answer content and the feedback information to obtain an updated knowledge base.

[0006] Optionally, obtaining question keywords based on the recording information includes: converting the recording information into recording text information; and extracting the question keywords from the recording text information.

[0007] Optionally, updating the knowledge base based on the response content and the feedback information to obtain an updated knowledge base includes: determining whether the target customer is satisfied with the response content to the question based on the feedback information; if the target customer is satisfied with the response content to the question, determining the first positive feedback keyword and the first negative feedback keyword in the feedback information; counting the number of the first positive feedback keyword and the number of the first negative feedback keyword; and updating the knowledge base based on the response content, the number of the first positive feedback keyword, and the number of the first negative feedback keyword to obtain the updated knowledge base.

[0008] Optionally, updating the knowledge base based on the response content, the number of the first positive feedback keywords, and the number of the first negative feedback keywords to obtain the updated knowledge base includes: determining the positive feedback ratio corresponding to the feedback information based on the number of the first positive feedback keywords and the number of the first negative feedback keywords; obtaining a first preset ratio; determining whether the positive feedback ratio is greater than or equal to the first preset ratio; and updating the knowledge base based on the response content if the positive feedback ratio is less than the first preset ratio to obtain the updated knowledge base.

[0009] Optionally, after determining whether the positive feedback ratio is greater than or equal to the first preset ratio, the method further includes: if the positive feedback ratio is greater than or equal to the first preset ratio, prohibiting the updating of the knowledge base.

[0010] Optionally, after determining whether the target customer is satisfied with the response to the question based on the feedback information, the method further includes: if the target customer is not satisfied with the response to the question, determining a second positive feedback keyword and a second negative feedback keyword in the feedback information; counting the number of the second positive feedback keywords and the number of the second negative feedback keywords; determining the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keywords and the number of the second negative feedback keywords; obtaining a second preset ratio; determining whether the negative feedback ratio is less than or equal to the second preset ratio; if the negative feedback ratio is less than or equal to the second preset ratio, updating the knowledge base based on the response content to obtain the updated knowledge base.

[0011] Optionally, after determining whether the back feedback ratio is less than or equal to the second preset ratio, the method further includes: if the back feedback ratio is greater than the second preset ratio, prohibiting the updating of the knowledge base.

[0012] Optionally, updating the knowledge base based on the response content to obtain the updated knowledge base includes: obtaining feedback information from each of M customers regarding the response content, where M is a positive integer; determining positive feedback keywords in the feedback information corresponding to each customer and counting the number of positive feedback keywords; determining the positive feedback ratio corresponding to the M customers based on the number of positive feedback keywords; obtaining a third preset ratio; determining whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio; and updating the knowledge base if the positive feedback ratio corresponding to the M customers is less than the third preset ratio to obtain the updated knowledge base.

[0013] Optionally, after determining whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio, the method further includes: if the positive feedback ratio corresponding to the M customers is greater than or equal to the third preset ratio, the knowledge base shall be prohibited from being updated.

[0014] According to another aspect of this application, a knowledge base updating device is also provided, comprising: a first acquisition module, configured to acquire recording information within a preset time period and a corresponding agent operation log, wherein the recording information is voice call information between a financial institution's agent and a target customer, and the agent operation log is an operation log in which the agent queries the answer content of a question from the knowledge base; a second acquisition module, configured to acquire question keywords based on the recording information, wherein the question keywords are keywords in a question raised by the target customer; a query module, configured to query the agent's operation record of the knowledge base from the agent operation log based on the question keywords, wherein the operation record includes at least a timestamp of the agent's query of the answer content of the question; a determination module, configured to determine the answer content of the question and the target customer's feedback information based on the timestamp and the recording information, wherein the feedback information is used to characterize the target customer's satisfaction with the answer content; and an update module, configured to update the knowledge base based on the answer content and the feedback information to obtain an updated knowledge base.

[0015] According to another aspect of this application, a computer-readable storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to execute the above-described method for updating the knowledge base at runtime.

[0016] According to another aspect of this application, an electronic device is also provided, the electronic device including one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are configured to run the programs, wherein the programs are configured to execute the above-described knowledge base update method during runtime.

[0017] In this application, firstly, audio recordings within a preset time period and corresponding agent operation logs are obtained. The audio recordings are voice calls between a financial institution's customer service representative and a target customer, and the agent operation logs are logs of the customer service representative's queries for answers to questions from a knowledge base. Then, question keywords are obtained based on the audio recordings, where the question keywords are keywords in the question raised by the target customer. Next, based on the question keywords, the agent operation logs are queried to record the agent's actions on the knowledge base, where the operation logs include at least a timestamp of the agent's query for the answer to the question. Then, based on the timestamp and the audio recordings, the answer to the question and the target customer's feedback are determined, where the feedback information represents the target customer's satisfaction with the answer. Finally, the knowledge base is updated based on the answer and the feedback information to obtain an updated knowledge base.

[0018] In the above process, keywords (i.e., question keywords) in the questions raised by target customers are automatically extracted from the voice call information (i.e., voice information) between the customer service representatives of financial institutions and target customers. Based on these keywords, the timestamps of the customer service representatives' responses to the questions are obtained from the agent operation logs. The responses and feedback information of the target customers are determined based on the timestamps and the recording information. This eliminates the need for business personnel to manually sample and analyze the call recordings, thus reducing labor and time costs. Then, the knowledge base is updated based on the responses and feedback information, resulting in an updated knowledge base. This further reduces labor and time costs and shortens the update time, thereby improving the technical effect of improving the update efficiency of the agent knowledge base. This solves the technical problem of low update efficiency of agent knowledge bases caused by manual updates in related technologies. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart of an optional knowledge base updating method according to an embodiment of this application;

[0021] Figure 2 This is a flowchart of another optional knowledge base updating method according to an embodiment of this application;

[0022] Figure 3 This is a schematic diagram of an optional knowledge base updating device according to an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] It should be noted that the knowledge base updating method, apparatus, storage medium and electronic device of this application can be used in the field of financial technology, as well as in other fields other than financial technology. The application fields of the knowledge base updating method, apparatus, storage medium and electronic device of this application are not limited.

[0027] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0028] Example 1

[0029] According to an embodiment of this application, an optional knowledge base update method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] Figure 1 This is a flowchart of an optional knowledge base updating method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0031] Step S101: Obtain the recording information within a preset time period and the corresponding agent operation log.

[0032] In one optional embodiment, a knowledge base update system can serve as the execution entity for the knowledge base update method in this application embodiment. For ease of description, the knowledge base update system will be referred to simply as the system below.

[0033] In step S101, the recorded information is the voice call information between the financial institution's customer service representative and the target customer, and the representative operation log is the operation log of the customer service representative querying the answer to the question from the knowledge base. In this embodiment, the knowledge base can be a representative knowledge base.

[0034] In this embodiment, the process can be automated by setting up RPA (Robotic Process Automation) (i.e. Figure 2 The system features a customized acquisition process, allowing it to automatically acquire recording information (i.e., based on timed, quantitative, or frequency-based acquisition) information. Figure 2 The system collects agent recording information and corresponding agent operation logs. For example, the system can transfer the recording files (i.e., recording information) to the system's RPA server via SFTP (Secure File Transfer Protocol), file sharing, or other methods.

[0035] Step S102: Obtain the keywords of the problem based on the recording information.

[0036] In step S102, the question keywords are the keywords in the questions raised by the target customer. These question keywords include, but are not limited to, keywords related to reporting a lost item, opening an account, and purchasing financial products.

[0037] In this embodiment, the system can convert recorded information into recorded text information (i.e., speech recognition technology) Figure 2 The analysis of agent recording information) and the extraction of question keywords from the recording text information (i.e. Figure 2 (Keywords for identifying problems).

[0038] Step S103: Based on the question keywords, query the agent's operation log for the operation records of the knowledge base.

[0039] In step S103, the operation log shall at least include a timestamp of the customer service representative's response to the inquiry. The operation log may also include the customer service representative's employee ID and the operation result, such as success, failure, or successful transfer.

[0040] For example, the system can search the agent operation log based on the keywords of the lost item reported in the audio text information, and find the record of the agent clicking on the knowledge base shortly after the target customer asked the question (e.g., within 5 seconds).

[0041] Step S104: Determine the response to the question and the feedback from the target customer based on the timestamp and the recording information.

[0042] In step S104, the feedback information is used to characterize the target customer's level of satisfaction with the response, such as whether the customer is satisfied with the response or dissatisfied with it.

[0043] In this embodiment, the system can determine the answer to the question and the feedback information of the target customer from the audio text information corresponding to the audio information based on the timestamp.

[0044] Step S105: Update the knowledge base based on the response content and feedback information to obtain the updated knowledge base.

[0045] For example, if the target customer is satisfied with the response, the system can update the knowledge base based on the number of first positive feedback keywords (e.g., "okay," "understand," "thank you," "I understand") and first negative feedback keywords (e.g., "incorrect," "don't understand," "cannot operate," "error report") in the feedback information, resulting in the updated knowledge base. If the target customer is dissatisfied with the response, the system can update the knowledge base based on the number of second positive feedback keywords and second negative feedback keywords in the feedback information, resulting in the updated knowledge base.

[0046] Optionally, the scheme defined in steps S101 to S105 above can be used to add content that does not exist in the knowledge base to the knowledge base. It can also be used to update existing but inaccurate content in the knowledge base. Furthermore, it can be used to delete existing content in the knowledge base whose related business has been discontinued.

[0047] Based on the scheme defined in steps S101 to S105 above, it can be understood that in this application, firstly, the recorded information within a preset time period and the corresponding agent operation log are obtained. The recorded information is the voice call information between a financial institution's customer service representative and the target customer, and the agent operation log is the operation log of the customer service representative querying the answer content of a question from the knowledge base. Then, question keywords are obtained based on the recorded information, where the question keywords are keywords in the question raised by the target customer. Next, based on the question keywords, the agent operation log is queried from the agent operation log to record the agent's operation on the knowledge base, where the operation record includes at least a timestamp of the agent's query for the answer content. Then, based on the timestamp and the recorded information, the answer content of the question and the target customer's feedback information are determined, where the feedback information is used to characterize the target customer's satisfaction with the answer content. Finally, the knowledge base is updated based on the answer content and the feedback information to obtain the updated knowledge base.

[0048] It should be noted that in the above process, by automatically extracting keywords (i.e., question keywords) from the voice call information (i.e., voice information) between the financial institution's customer service representatives and the target customer, and obtaining the timestamp of the customer service representative's response to the question from the agent's operation log based on these keywords, the response content and the target customer's feedback information are determined based on the timestamp and the recording information. This eliminates the need for business personnel to manually sample and analyze the call recordings to analyze the customer's question, the agent's response content, and the customer's feedback information, thus reducing labor and time costs. Then, the knowledge base is updated based on the response content and feedback information to obtain an updated knowledge base. This eliminates the need for business personnel to manually update the knowledge base content, further reducing labor and time costs and the update time. This achieves the technical effect of improving the update efficiency of the agent knowledge base, thereby solving the technical problem of low update efficiency of the agent knowledge base caused by manual updates in related technologies.

[0049] Optionally, in the knowledge base update method provided in this application embodiment, obtaining question keywords based on recording information includes: the system can convert the recording information into recording text information; and then extract question keywords from the recording text information.

[0050] In this embodiment, the system can convert recorded information into recorded text information using speech recognition technology, and extract question keywords from the recorded text information.

[0051] Optionally, in the knowledge base update method provided in this application embodiment, the knowledge base is updated based on the response content and feedback information to obtain an updated knowledge base, including: the system can determine whether the target customer is satisfied with the response content to the question based on the feedback information; if the target customer is satisfied with the response content to the question, the system can determine the first positive feedback keyword and the first negative feedback keyword in the feedback information; then count the number of the first positive feedback keyword and the number of the first negative feedback keyword; finally, the knowledge base is updated based on the response content, the number of the first positive feedback keyword, and the number of the first negative feedback keyword to obtain an updated knowledge base.

[0052] In this embodiment, the system can determine whether the target customer is satisfied with the answer to the question based on the feedback information. If the target customer is satisfied with the answer, the system updates the knowledge base based on the answer content, the number of first positive feedback keywords in the feedback information, and the number of first negative feedback keywords in the feedback information, thus obtaining an updated knowledge base. For example, when the system determines that the target customer is satisfied with the answer to the question based on the feedback information, it can determine the positive feedback ratio of the target customer through the number of first positive feedback keywords and the number of first negative feedback keywords, and update the knowledge base based on the answer content and the positive feedback ratio.

[0053] It should be noted that by automatically determining whether the target customer is satisfied with the answer to the question based on the feedback information, and when the target customer is satisfied with the answer, the knowledge base is automatically updated based on the answer content, the number of first positive feedback keywords in the feedback information, and the number of first negative feedback keywords in the feedback information. This solves the problem of low update efficiency of the agent knowledge base caused by manual updates, and thus improves the update efficiency of the agent knowledge base.

[0054] Optionally, in the knowledge base update method provided in this application embodiment, the knowledge base is updated based on the response content, the number of first positive feedback keywords, and the number of first negative feedback keywords to obtain an updated knowledge base. This includes: the system can determine the positive feedback ratio corresponding to the feedback information based on the number of first positive feedback keywords and the number of first negative feedback keywords; then obtain a first preset ratio; then determine whether the positive feedback ratio is greater than or equal to the first preset ratio; finally, if the positive feedback ratio is less than the first preset ratio, the knowledge base is updated based on the response content to obtain an updated knowledge base.

[0055] Optionally, in the knowledge base update method provided in the embodiments of this application, after determining whether the positive feedback ratio is greater than or equal to the first preset ratio, the method further includes: if the positive feedback ratio is greater than or equal to the first preset ratio, the system may prohibit updating the knowledge base.

[0056] In this embodiment, the first preset ratio is a pre-set ratio value corresponding to the positive feedback ratio. For example... Figure 2 As shown, the system can determine the positive feedback ratio corresponding to the feedback information based on the number of first positive feedback keywords and the number of first negative feedback keywords, and determine whether to update the knowledge base by judging whether the positive feedback ratio is greater than or equal to a first preset ratio. The specific expression is as follows:

[0057] Do not update the knowledge base

[0058] Update the knowledge base

[0059] Wherein, W1 is the positive feedback ratio, S1 is the number of positive feedback samples, K1 is the number of first positive feedback keywords, S2 is the number of negative feedback samples, K2 is the number of first negative feedback keywords, and R1 is the first preset ratio.

[0060] For example, when the positive feedback ratio is 40% and the first preset ratio is 50%, the system can update the knowledge base based on the response content to obtain the updated knowledge base; when the positive feedback ratio is 80% and the first preset ratio is 50%, updating the knowledge base is prohibited.

[0061] It should be noted that by determining the positive feedback ratio corresponding to the feedback information based on the number of first positive feedback keywords and the number of first negative feedback keywords, and by automatically judging whether the positive feedback ratio is greater than or equal to the first preset ratio to determine whether to update the knowledge base, the agent knowledge base can be updated in a timely manner when customer satisfaction is low, thereby improving the update efficiency of the agent knowledge base.

[0062] Optionally, in the knowledge base update method provided in this application embodiment, after determining whether the target customer is satisfied with the answer to the question based on the feedback information, the method further includes: if the target customer is not satisfied with the answer to the question, the system can determine the second positive feedback keywords and the second negative feedback keywords in the feedback information; then count the number of the second positive feedback keywords and the number of the second negative feedback keywords; then determine the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keywords and the number of the second negative feedback keywords; obtain a second preset ratio; then determine whether the negative feedback ratio is less than or equal to the second preset ratio; if the negative feedback ratio is less than or equal to the second preset ratio, the system can update the knowledge base based on the answer content to obtain the updated knowledge base.

[0063] Optionally, in the knowledge base update method provided in the embodiments of this application, after determining whether the reverse feedback ratio is less than or equal to the second preset ratio, the method further includes: if the reverse feedback ratio is greater than the second preset ratio, the system may prohibit updating the knowledge base.

[0064] In this embodiment, the second preset ratio is a pre-set ratio value corresponding to the reverse feedback ratio. For example... Figure 2 As shown, if the target customer is dissatisfied with the response to the question, the system can determine the corresponding reverse feedback ratio based on the number of second positive feedback keywords and the number of second negative feedback keywords, and determine whether to update the knowledge base by judging whether the reverse feedback ratio is greater than or equal to the first preset ratio. The specific expression is as follows:

[0065] Do not update the knowledge base

[0066] Update the knowledge base

[0067] Wherein, W2 is the reverse feedback ratio, S3 is the number of positive feedback samples, K3 is the number of second positive feedback keywords, S4 is the number of reverse feedback samples, K4 is the number of second reverse feedback keywords, and R2 is the second preset ratio.

[0068] For example, when the negative feedback ratio is 60% and the second preset ratio is 50%, the system can update the knowledge base based on the response content to obtain the updated knowledge base; when the positive feedback ratio is 30% and the second preset ratio is 50%, updating the knowledge base is prohibited.

[0069] It should be noted that by determining the reverse feedback ratio corresponding to the feedback information based on the number of second positive feedback keywords and the number of second negative feedback keywords, and by automatically judging whether the reverse feedback ratio is less than or equal to the second preset ratio to determine whether to update the knowledge base, the agent knowledge base can be updated in a timely manner when customer satisfaction is low, thereby improving the update efficiency of the agent knowledge base.

[0070] Optionally, in the knowledge base update method provided in this application embodiment, updating the knowledge base based on the response content to obtain an updated knowledge base includes: the system can obtain feedback information on the response content from each of M customers, where M is a positive integer; then determine the positive feedback keywords in the feedback information corresponding to each customer, and count the number of positive feedback keywords; then determine the positive feedback ratio corresponding to the M customers based on the number of positive feedback keywords; obtain a third preset ratio; then determine whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio; if the positive feedback ratio corresponding to the M customers is less than the third preset ratio, the system can update the knowledge base to obtain an updated knowledge base.

[0071] Optionally, in the knowledge base update method provided in the embodiments of this application, after determining whether the positive feedback ratio corresponding to M customers is less than a third preset ratio, the method further includes: if the positive feedback ratio corresponding to M customers is greater than or equal to the third preset ratio, the system may prohibit updating the knowledge base.

[0072] In this embodiment, the third preset ratio is a pre-set ratio value corresponding to the positive feedback ratio for M customers. For example... Figure 2As shown, the system can obtain feedback information from each of the M customers regarding the response content, determine the positive feedback ratio for the M customers based on the number of positive feedback keywords in each customer's feedback information, and then determine whether to update the knowledge base by judging whether the positive feedback ratio for the M customers is less than a third preset ratio. The specific expression is as follows:

[0073] Do not update the knowledge base

[0074] Update the knowledge base

[0075] Where S12 is the positive feedback keyword, K12 is the number of positive feedback samples, S is the total number of samples, and R12 is the third preset ratio.

[0076] Optionally, the system can update the knowledge base by analyzing the feedback information from each customer. Specifically, the system can determine whether the response content does not exist in the knowledge base; if the response content does not exist in the knowledge base, the system obtains new response content corresponding to the question raised by the target customer and adds the new response content to the knowledge base. If the response content already exists in the knowledge base, the system can determine whether the response content is accurate; if the response content is inaccurate, the system obtains new response content corresponding to the question raised by the target customer and adds the new response content to the knowledge base.

[0077] Optionally, if the response content does not exist in the knowledge base, the system can determine whether the business related to the question has been taken offline; if the business related to the question has been taken offline, the system will delete the response content from the knowledge base.

[0078] Therefore, the knowledge base update method proposed in this application can automatically extract keywords (i.e., question keywords) from the voice call information (i.e., voice information) between the customer service representatives of financial institutions and the target customer. Based on these keywords, the timestamps of the customer service representatives' responses to the questions are obtained from the agent operation logs. The responses and customer feedback are then determined based on the timestamps and the recording information. This eliminates the need for manual sampling and analysis of the call recordings by business personnel, reducing labor and time costs. The knowledge base is then updated based on the responses and feedback, resulting in an updated knowledge base. This further reduces labor and time costs and shortens the update time, thereby improving the efficiency of updating the agent knowledge base. This solves the technical problem of low update efficiency in related technologies that rely on manual updates of the agent knowledge base.

[0079] Example 2

[0080] According to an embodiment of this application, an embodiment of a knowledge base updating device is provided, wherein, Figure 3 This is a schematic diagram of an optional knowledge base updating device according to an embodiment of this application, such as... Figure 3 As shown, the device includes:

[0081] The first acquisition module 301 is used to acquire the recording information within a preset time period and the corresponding agent operation log. The recording information is the voice call information between the agent of the financial institution and the target customer, and the agent operation log is the operation log of the agent querying the answer content of the question from the knowledge base.

[0082] The second acquisition module 302 is used to acquire question keywords based on the recording information, wherein the question keywords are the keywords in the questions raised by the target customer;

[0083] The query module 303 is used to query the operation records of customer service representatives on the knowledge base from the agent operation log based on the question keywords. The operation records include at least the timestamp of the customer service representative's reply to the question.

[0084] The determination module 304 is used to determine the answer content of the question and the feedback information of the target customer based on the timestamp and the recording information, wherein the feedback information is used to characterize the target customer's satisfaction with the answer content;

[0085] The update module 305 is used to update the knowledge base based on the response content and feedback information to obtain the updated knowledge base.

[0086] It should be noted that the first acquisition module 301, the second acquisition module 302, the query module 303, the determination module 304, and the update module 305 mentioned above correspond to steps S101 to S105 in the above embodiment 1. The five modules and the corresponding steps implement the same examples and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0087] Optionally, the second acquisition module includes: a conversion unit for converting the recording information into recording text information; and an extraction unit for extracting question keywords from the recording text information.

[0088] Optionally, the update module includes: a first judgment unit, used to judge whether the target customer is satisfied with the answer to the question based on the feedback information; a first determination unit, used to determine the first positive feedback keyword and the first negative feedback keyword in the feedback information if the target customer is satisfied with the answer to the question; a first statistics unit, used to count the number of the first positive feedback keyword and the number of the first negative feedback keyword; and a first update unit, used to update the knowledge base based on the answer content, the number of the first positive feedback keyword, and the number of the first negative feedback keyword, to obtain the updated knowledge base.

[0089] Optionally, the first updating unit includes: a second determining unit, configured to determine the positive feedback ratio corresponding to the feedback information based on the number of first positive feedback keywords and the number of first negative feedback keywords; a first obtaining unit, configured to obtain a first preset ratio; a second judging unit, configured to judge whether the positive feedback ratio is greater than or equal to the first preset ratio; and a second updating unit, configured to update the knowledge base based on the response content when the positive feedback ratio is less than the first preset ratio, thereby obtaining an updated knowledge base.

[0090] Optionally, the knowledge base update device further includes: a first update prohibition unit, used to prohibit updating the knowledge base if the positive feedback ratio is greater than or equal to the first preset ratio after determining whether the positive feedback ratio is greater than or equal to the first preset ratio.

[0091] Optionally, the knowledge base updating device further includes: a third determining unit, used to determine the second positive feedback keywords and the second negative feedback keywords in the feedback information when the target customer is dissatisfied with the answer to the question; a second statistical unit, used to count the number of the second positive feedback keywords and the number of the second negative feedback keywords; a fourth determining unit, used to determine the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keywords and the number of the second negative feedback keywords; a second obtaining unit, used to obtain a second preset ratio; a third judging unit, used to judge whether the negative feedback ratio is less than or equal to the second preset ratio; and a third updating unit, used to update the knowledge base according to the answer content when the negative feedback ratio is less than or equal to the second preset ratio, to obtain the updated knowledge base.

[0092] Optionally, the knowledge base update device further includes: a second update prohibition unit, used to prohibit updating the knowledge base if the back feedback ratio is greater than the second preset ratio after determining whether the back feedback ratio is less than or equal to the second preset ratio.

[0093] Optionally, the knowledge base updating device further includes: a third acquisition unit, used to acquire feedback information of each of the M customers on the response content, where M is a positive integer; a fifth determination unit, used to determine the positive feedback keywords in the feedback information corresponding to each customer and count the number of positive feedback keywords; a sixth determination unit, used to determine the positive feedback ratio corresponding to the M customers based on the number of positive feedback keywords; a fourth acquisition unit, used to acquire a third preset ratio; a fourth judgment unit, used to judge whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio; and a fourth update unit, used to update the knowledge base when the positive feedback ratio corresponding to the M customers is less than the third preset ratio, to obtain the updated knowledge base.

[0094] Optionally, the third prohibition of update unit is used to prohibit updating the knowledge base if the positive feedback ratio corresponding to M customers is greater than or equal to the third preset ratio after determining whether the positive feedback ratio corresponding to M customers is less than the third preset ratio.

[0095] Example 3

[0096] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to execute the above-described knowledge base update method at runtime.

[0097] Example 4

[0098] According to another aspect of the embodiments of this application, an electronic device is also provided, wherein, Figure 4This is a schematic diagram of an optional electronic device according to an embodiment of this application, such as... Figure 4 As shown, the electronic device includes one or more processors; and a memory for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the program for running, wherein the program is configured to execute the aforementioned knowledge base update method during runtime.

[0099] like Figure 4 As shown, this application embodiment provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps:

[0100] The system acquires audio recordings within a preset time period, along with corresponding agent operation logs. The audio recordings are voice calls between a financial institution's customer service representatives and the target customer. The agent operation logs are logs of the customer service representatives' queries of the knowledge base for answers to questions. Based on the audio recordings, the system obtains question keywords, which are keywords from the questions raised by the target customer. Based on the question keywords, the system queries the agent operation logs for the customer service representatives' operations on the knowledge base. These operation logs include at least a timestamp of the customer service representatives' queries for answers to questions. Based on the timestamps and audio recordings, the system determines the answer to the question and the target customer's feedback, which indicates the target customer's satisfaction with the answer. Finally, the system updates the knowledge base based on the answer and feedback, resulting in an updated knowledge base.

[0101] Optionally, the processor may also perform the following steps when executing the program: converting the recording information into recording text information; and extracting question keywords from the recording text information.

[0102] Optionally, when the processor executes the program, it also performs the following steps: determining whether the target customer is satisfied with the answer to the question based on the feedback information; if the target customer is satisfied with the answer to the question, determining the first positive feedback keyword and the first negative feedback keyword in the feedback information; counting the number of the first positive feedback keyword and the number of the first negative feedback keyword; updating the knowledge base based on the answer content, the number of the first positive feedback keyword, and the number of the first negative feedback keyword to obtain the updated knowledge base.

[0103] Optionally, when the processor executes the program, it also performs the following steps: determining the positive feedback ratio corresponding to the feedback information based on the number of first positive feedback keywords and the number of first negative feedback keywords; obtaining a first preset ratio; determining whether the positive feedback ratio is greater than or equal to the first preset ratio; and updating the knowledge base based on the response content when the positive feedback ratio is less than the first preset ratio, thereby obtaining the updated knowledge base.

[0104] Optionally, the processor also performs the following steps when executing the program: if the positive feedback ratio is greater than or equal to a first preset ratio, the knowledge base is prohibited from being updated.

[0105] Optionally, the processor, when executing the program, further implements the following steps: after determining whether the target customer is satisfied with the answer to the question based on the feedback information, if the target customer is not satisfied with the answer to the question, determine the second positive feedback keyword and the second negative feedback keyword in the feedback information; count the number of the second positive feedback keyword and the number of the second negative feedback keyword; determine the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keyword and the number of the second negative feedback keyword; obtain a second preset ratio; determine whether the negative feedback ratio is less than or equal to the second preset ratio; if the negative feedback ratio is less than or equal to the second preset ratio, update the knowledge base based on the answer content to obtain the updated knowledge base.

[0106] Optionally, when the processor executes the program, it also performs the following steps: after determining whether the reverse feedback ratio is less than or equal to the second preset ratio, if the reverse feedback ratio is greater than the second preset ratio, it prohibits updating the knowledge base.

[0107] Optionally, the processor, when executing the program, also implements the following steps: obtaining feedback information from each of the M customers regarding the response content, where M is a positive integer; determining the positive feedback keywords in the feedback information corresponding to each customer and counting the number of positive feedback keywords; determining the positive feedback ratio corresponding to the M customers based on the number of positive feedback keywords; obtaining a third preset ratio; determining whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio; and updating the knowledge base if the positive feedback ratio corresponding to the M customers is less than the third preset ratio, thereby obtaining the updated knowledge base.

[0108] Optionally, the processor also performs the following steps when executing the program: after determining whether the positive feedback ratio corresponding to M customers is less than the third preset ratio, if the positive feedback ratio corresponding to M customers is greater than or equal to the third preset ratio, the knowledge base is prohibited from being updated.

[0109] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0110] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0112] The units described as separate components may or may not be physically separate. Similarly, the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0113] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0114] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0115] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of updating a knowledge base, characterized by, include: Acquire audio recordings within a preset time period and the corresponding agent operation logs, wherein the audio recordings are voice call information between a financial institution's customer service representative and a target customer, and the agent operation logs are operation logs of the customer service representative querying answers to questions from a knowledge base. Based on the recorded information, question keywords are obtained, wherein the question keywords are keywords in the questions raised by the target customer; Based on the keywords of the question, query the operation records of the customer service representative on the knowledge base from the operation log, wherein the operation records include at least the timestamp of the customer service representative's query for the answer to the question; The response to the question and the feedback information of the target customer are determined based on the timestamp and the recording information, wherein the feedback information is used to characterize the target customer's satisfaction with the response. The knowledge base is updated based on the response content and the feedback information to obtain an updated knowledge base, including: determining whether the target customer is satisfied with the response content to the question based on the feedback information; if the target customer is not satisfied with the response content to the question, identifying a second positive feedback keyword and a second negative feedback keyword in the feedback information; counting the number of the second positive feedback keywords and the number of the second negative feedback keywords; determining the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keywords and the number of the second negative feedback keywords; obtaining a second preset ratio; determining whether the negative feedback ratio is less than or equal to the second preset ratio; and updating the knowledge base based on the response content if the negative feedback ratio is less than or equal to the second preset ratio to obtain the updated knowledge base.

2. The method of claim 1, wherein, Based on the recorded information, the question keywords are obtained, including: Convert the recorded information into recorded text information; Extract the keywords related to the problem from the audio recording text.

3. The method according to claim 1, characterized in that, After determining whether the target customer is satisfied with the response to the question based on the feedback information, the method further includes: If the target customer is satisfied with the response to the question, determine the first positive feedback keyword and the first negative feedback keyword in the feedback information; Count the number of the first positive feedback keywords and the number of the first negative feedback keywords; The knowledge base is updated based on the content of the response, the number of the first positive feedback keywords, and the number of the first negative feedback keywords to obtain the updated knowledge base.

4. The method according to claim 3, characterized in that, The knowledge base is updated based on the response content, the number of the first positive feedback keywords, and the number of the first negative feedback keywords to obtain the updated knowledge base, including: The positive feedback ratio corresponding to the feedback information is determined based on the number of the first positive feedback keywords and the number of the first negative feedback keywords; Obtain the first preset ratio; Determine whether the positive feedback ratio is greater than or equal to the first preset ratio; If the positive feedback ratio is less than the first preset ratio, the knowledge base is updated based on the response content to obtain the updated knowledge base.

5. The method according to claim 4, characterized in that, After determining whether the positive feedback ratio is greater than or equal to the first preset ratio, the method further includes: If the positive feedback ratio is greater than or equal to the first preset ratio, updating the knowledge base is prohibited.

6. The method according to claim 1, characterized in that, After determining whether the reverse feedback ratio is less than or equal to the second preset ratio, the method further includes: If the reverse feedback ratio is greater than the second preset ratio, updating the knowledge base is prohibited.

7. The method according to claim 1 or claim 4, characterized in that, The knowledge base is updated based on the response content to obtain the updated knowledge base, including: Obtain feedback information from each of the M customers regarding the response content, where M is a positive integer; Identify the positive feedback keywords in the feedback information corresponding to each customer, and count the number of the positive feedback keywords; The positive feedback ratio for each of the M customers is determined based on the number of positive feedback keywords. Obtain the third preset ratio; Determine whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio; If the proportion of positive feedback corresponding to the M customers is less than the third preset proportion, the knowledge base is updated to obtain the updated knowledge base.

8. The method according to claim 7, characterized in that, After determining whether the positive feedback ratio corresponding to the M customers is less than the third preset ratio, the method further includes: If the positive feedback ratio corresponding to the M customers is greater than or equal to the third preset ratio, updating the knowledge base is prohibited.

9. A knowledge base updating device, characterized in that, include: The first acquisition module is used to acquire recording information within a preset time period and the corresponding agent operation log of the recording information. The recording information is voice call information between the agent of the financial institution and the target customer, and the agent operation log is the operation log of the agent querying the answer content of the question from the knowledge base. The second acquisition module is used to acquire question keywords based on the recording information, wherein the question keywords are keywords in the questions raised by the target customer; The query module is used to query the operation records of the customer service representative on the knowledge base from the agent operation log based on the question keywords, wherein the operation records include at least the timestamp of the customer service representative's reply to the question; The determining module is used to determine the answer content of the question and the feedback information of the target customer based on the timestamp and the recording information, wherein the feedback information is used to characterize the target customer's satisfaction with the answer content; An update module is used to update the knowledge base based on the response content and the feedback information to obtain an updated knowledge base; wherein, the update module includes: a first judgment unit, used to judge whether the target customer is satisfied with the response content to the question based on the feedback information; The third determining unit is used to determine the second positive feedback keyword and the second negative feedback keyword in the feedback information when the target customer is not satisfied with the response to the question; the second statistical unit is used to count the number of the second positive feedback keyword and the number of the second negative feedback keyword; the fourth determining unit is used to determine the negative feedback ratio corresponding to the feedback information based on the number of the second positive feedback keyword and the number of the second negative feedback keyword; the second obtaining unit is used to obtain the second preset ratio; the third judging unit is used to judge whether the negative feedback ratio is less than or equal to the second preset ratio; the third updating unit is used to update the knowledge base according to the response content when the negative feedback ratio is less than or equal to the second preset ratio, to obtain the updated knowledge base.

10. A computer-readable storage medium, characterized in that, A computer-readable storage medium stores a computer program, wherein the computer program is configured to execute, at runtime, the method for updating the knowledge base as described in any one of claims 1 to 8.

11. An electronic device, characterized in that, It includes one or more processors and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the knowledge base update method according to any one of claims 1 to 8.

Images