Session management method and device, computing equipment, storage medium and program product

Abstract

The invention provides a session management method for a customer service scene, and the method comprises the steps: responding to a received session initiation operation of an object, and transmitting a session initiation request comprising an object identifier to a server; receiving feedback information indicating that a target customer service is determined aiming at the session initiation request from the server, wherein the target customer service is a customer service which is determined based on the object identifier and matched with the object; in response to the received session input operation of the object, session information is sent to a server; in response to reply information received from the server, the reply information is presented, and the reply information is generated by the target customer service based on the session information; and in response to the object exiting the current session, sending information indicating the ending of the current session to the server, so that the server deletes at least a part of session records of the current session when receiving the information indicating the ending of the current session and determining that a session deletion condition is satisfied. Through the method, the session data security of the customer service scene can be improved, and the session experience is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to a conversation management method and apparatus for customer service scenarios, a computing device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Thanks to the development of computer technology and the widespread use of the internet, people are increasingly using computers and networks to conduct various activities. This has greatly facilitated people's lives, studies, and work, but it has also brought many risks, such as the highly-publicized risk of data leakage. For example, online customer service can be provided via the internet, allowing people to quickly and easily resolve various issues or seek relevant services by communicating with online customer service. However, this online communication process often involves relatively sensitive or private information and data. Preventing the leakage of such information and data, ensuring data security, and enhancing people's trust in online customer service have become urgent issues that need to be addressed. Summary of the Invention

[0003] In view of this, the present disclosure provides a session management method and apparatus, a computing device, a computer-readable storage medium, and a computer program product for customer service scenarios, which can alleviate, mitigate, or even eliminate the above-mentioned problems.

[0004] According to one aspect of the present disclosure, a session management method for a customer service scenario is provided, comprising: in response to receiving a session initiation operation of an object, sending a session initiation request to a server, the session initiation request being used to initiate a current conversation and including an object identifier of an object using a client; receiving feedback information from the server regarding the session initiation request, the feedback information being configured to indicate that a target customer service has been determined, wherein the target customer service is a customer service that matches the object determined based on the object identifier; in response to receiving a session input operation of the object, sending session information to the server, the session information including session content input by the object through the session input operation; in response to receiving reply information from the server, presenting reply information, wherein the reply information is generated by the target customer service based on the session information; and in response to the object exiting the current session, sending information indicating the end of the current session to the server, so that upon receiving the information indicating the end of the current session and determining that a session deletion condition is satisfied, the server deletes at least a portion of the session record of the current session, the session deletion condition including at least one of the following: the current session involves private data, the current session belongs to a private session scenario, and a deletion instruction from the object for at least a portion of the session record of the current session is received.

[0005] In some embodiments, the above method further includes: presenting a watermark in a session interface, where the session interface is an interface for presenting a current session, and the watermark includes at least one of the following: an object identifier, a timestamp, and a session identifier.

[0006] According to another aspect of the present disclosure, a session management method for a customer service scenario is provided, comprising: receiving a session initiation request from a client, the session initiation request being used to initiate a current conversation and including an object identifier of an object using the client; determining a target customer service matching the object based on the object identifier; sending feedback information regarding the session initiation request to the client, the feedback information being configured to indicate that the target customer service has been determined; receiving session information from the client, the session information including session content input by the object through the client; generating, by the target customer service, and sending a reply message to the client based on the received session information; and deleting at least a portion of the session record of the current session in response to the current session ending and satisfying a session deletion condition, the session deletion condition including at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction from the object for at least a portion of the session record of the current session is received.

[0007] In some embodiments, the above method further includes: sending session deletion information to the client, where the session deletion information is used to indicate that at least a portion of the session record of the current session has been deleted.

[0008] According to another aspect of the present disclosure, a session management device for a customer service scenario is provided, comprising: a session initiation module, configured to: in response to receiving a session initiation operation of an object, send a session initiation request to a server, the session initiation request being used to initiate a current conversation and including an object identifier of an object using a client; a receiving module, configured to: receive feedback information from the server regarding the session initiation request, the feedback information being configured to indicate that a target customer service has been determined, wherein the target customer service is a customer service that matches the object based on the object identifier; a first sending module, configured to: in response to receiving a session input operation of the object, send session information to the server, the session information including information about the object through the session input operation The first embodiment comprises a first sending module, a second sending module, and a second sending module. The first sending module is configured to: present the reply information in response to receiving the reply information from the server, wherein the reply information is generated by the target customer service based on the session information; the second sending module is configured to: send information indicating the end of the current session to the server in response to the object exiting the current session, so that the server deletes at least part of the session record of the current session when receiving the information indicating the end of the current session and determining that the session deletion condition is met, and the session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction for at least part of the session record of the current session is received from the object.

[0009] In some embodiments, the feedback information includes customer service characteristic data corresponding to the target customer service, and the above-mentioned device also includes a customer service presentation module, which is configured to: present a virtual customer service image corresponding to the target customer service based on the customer service characteristic data.

[0010] In some embodiments, the above-mentioned device also includes a screenshot protection module, which is configured to: present a privacy protection option in response to detecting a screenshot operation for the current session; and perform blurring processing on the privacy data in the screenshot image corresponding to the screenshot operation in response to receiving a confirmation operation for the privacy protection option.

[0011] In some embodiments, the screenshot protection module is further configured to: perform blurring processing on the privacy data in the screenshot image corresponding to the screenshot operation, including: identifying target keywords in the screenshot image, where the target keywords are pre-set keywords related to the privacy data; determining the privacy data area based on the position of the target keywords in the screenshot image; and performing blurring processing on the privacy data area in the screenshot image.

[0012] In some embodiments, the screenshot protection module is further configured to: extract the private data area in the screenshot image; filter the extracted private data area to obtain a blurred private data area; and merge the blurred private data area with the screenshot image to obtain a screenshot image with blurred private data.

[0013] In some embodiments, the apparatus further includes a deletion module configured to delete at least a portion of the session record of the current session in response to the object exiting the current session and satisfying a session deletion condition.

[0014] In some embodiments, the apparatus further includes a verification module configured to: present a password verification control in response to receiving a processing operation for at least a portion of the locally stored session record; receive a verification password input by the subject through the password verification control; and process at least a portion of the session record according to the processing operation in response to the verification password matching a preset password. The preset password is a password pre-set by the subject to protect at least a portion of the locally stored session record.

[0015] In some embodiments, the above-mentioned device also includes a watermark module, which is configured to: present a watermark in a session interface, where the session interface is an interface for presenting the current session, and the watermark includes at least one of the following items: object identifier, timestamp, and session identifier.

[0016] In some embodiments, the first sending module is further configured to: encrypt content input by the subject through the session input operation according to a first key to obtain session information, and send the session information to the server; and wherein the presenting module is further configured to: decrypt the reply information according to a second key and present the decrypted reply information. The first key and the second key are pre-set keys used for encrypted transmission of the current session.

[0017] According to another aspect of the present disclosure, a session management device for a customer service scenario is provided, comprising: a first receiving module configured to receive a session initiation request from a client, the session initiation request being used to initiate a current conversation and including an object identifier of an object using the client; a determination module configured to determine, based on the object identifier, a target customer service matching the object; a sending module configured to send feedback information regarding the session initiation request to the client, the feedback information being configured to indicate that the target customer service has been determined; a second receiving module configured to receive session information from the client, the session information including session content input by the object through the client; a reply module configured to generate and send a reply message to the client via the target customer service based on the received session information; and a deletion module configured to delete, in response to the current session ending and satisfying a session deletion condition, at least a portion of the session record of the current session, the session deletion condition including at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction from the object for at least a portion of the session record of the current session is received.

[0018] In some embodiments, the determination module is further configured to: obtain feature data corresponding to the object based on the object identification, the feature data representing at least one of the object behavior and object attributes of the object; and determine a target customer service that matches the object based on the feature data.

[0019] In some embodiments, the determination module is further configured to: determine the similarity between the characteristic data and the characteristic data of each alternative customer service among multiple alternative customer service personnel, wherein the characteristic data of each alternative customer service personnel is the characteristic data of an object that matches the alternative customer service personnel; and determine the target customer service personnel based on the alternative customer service personnel with the highest similarity.

[0020] In some embodiments, the determination module is further configured to determine the target customer service by adjusting the conversation style of the candidate customer service with the highest similarity based on the feature data to match the conversation style preferred by the object reflected by the feature data.

[0021] In some embodiments, the above-mentioned device also includes an evaluation module, which is configured to: obtain the subject's conversation evaluation of the current conversation; in response to the conversation evaluation meeting a preset evaluation threshold, determine a new alternative customer service or update the corresponding alternative customer service based on the target customer service.

[0022] In some embodiments, the sending module is further configured to: send session deletion information to the client, where the session deletion information is used to indicate that at least a portion of the session record of the current session has been deleted.

[0023] In some embodiments, the reply module is further configured to: decrypt the received session information using a first key; generate a reply message through the target customer service based on the decrypted session information; encrypt the generated reply message using a second key to obtain a reply message; and send the reply message to the client. The first key and the second key are pre-set keys used for encrypted transmission of the current session.

[0024] In some embodiments, the reply module is further configured to: obtain at least a portion of the session record of the historical session in response to the session information containing information related to the historical session; and generate and send a reply message to the client through the target customer service based on at least a portion of the session record and session information of the historical session.

[0025] According to another aspect of the present disclosure, a computing device is provided, comprising: a memory configured to store computer-executable instructions; and a processor configured to perform the method provided according to the aforementioned aspect when the computer-executable instructions are executed by the processor.

[0026] According to another aspect of the present disclosure, a computer-readable storage medium is provided, which stores computer-executable instructions. When the computer-executable instructions are executed, the method provided according to the above aspects is executed.

[0027] According to yet another aspect of the present disclosure, a computer program product is provided, comprising computer executable instructions, which implement the method provided according to the aforementioned aspects when the computer instructions are executed by a processor.

[0028] Through the session management solution for customer service scenarios provided by the present disclosure, the object can send a session initiation request to the server through the client to initiate the current session. The server can determine the target customer service that matches the object based on the object identifier. The object can then communicate online with the target customer service through the client. When the object exits the current session, the server can delete at least a portion of the session record of the current session if it determines that the current session involves private data, the current session belongs to a private session scenario, or receives a deletion instruction from the object for at least a portion of the session record of the current session. In this way, after the object exits the current session, various types of relatively private data or data that the object does not want to save can be deleted in a timely manner, thereby avoiding such data from being retained in the server. This can greatly reduce the risk of leakage of such data and improve data security. At the same time, this also helps to improve the object's trust in customer service, dispel or reduce the object's concerns about privacy leakage, ensure the safe and efficient development of customer service, and improve the object's satisfaction with customer service.

[0029] These and other aspects of the disclosure will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Further details, features and advantages of the present disclosure are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0031] Figure 1 Schematically illustrates example scenarios in which the technical solutions provided by some embodiments of the present disclosure can be applied;

[0032] Figure 2 Schematically illustrates an example flow chart of a session management method for a customer service scenario according to some embodiments of the present disclosure;

[0033] Figure 3 Another example flow chart of a session management method for a customer service scenario according to some embodiments of the present disclosure is schematically shown;

[0034] Figure 4 An example interaction diagram of a session management method for a customer service scenario according to some embodiments of the present disclosure is schematically shown;

[0035] Figure 5 Schematically illustrates an example architecture diagram of a client for executing a session management method according to some embodiments of the present disclosure;

[0036] Figure 6 Schematically illustrates an example architecture diagram of a server side for executing a session management method according to some embodiments of the present disclosure;

[0037] Figure 7 Schematically illustrates an example block diagram of a session management device for customer service scenarios according to some embodiments of the present disclosure;

[0038] Figure 8 Schematically illustrates another example block diagram of a session management device for a customer service scenario according to some embodiments of the present disclosure;

[0039] Figure 9 An example block diagram of a computing device according to some embodiments of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0040] Before introducing the embodiments of the present disclosure in detail, some related concepts are first explained.

[0041] In the embodiments of the present disclosure, "customer service" can be understood as an object that provides customer service. Customer service can include answering questions raised by customers, solving problems encountered by customers, and other content related to improving customer satisfaction. Customer service can include manual customer service and electronic customer service. Among them, electronic customer service can be implemented with the help of pre-written programs, algorithms, or pre-trained machine learning models. In addition, the customer service scenario mentioned in the embodiments of the present disclosure can be understood as a scenario in which an object interacts with customer service. The interaction method can be text interaction, picture interaction, voice interaction, video interaction, etc., or it can also be a combination of multiple interaction methods.

[0042] In the embodiments of the present disclosure, an "object" can be understood as any type of customer object that can conduct a conversation with customer service. It can be an object of the natural person type, or an object of various organizational types such as a company, association, etc., or a robot or other form of object. In some embodiments of the present disclosure, an object can have a unique identifier, such as a UIN (User Identification Number). In other words, in some embodiments, different objects can be distinguished by different identifiers. Exemplarily, the identifier of an object can be the name, number, graphic code, or other form of identifier of the object.

[0043] In the embodiments of the present disclosure, the description of the terms "one embodiment", "another embodiment", "some embodiments", etc. should be understood as that the specific features, structures or steps described in conjunction with the embodiment are included in at least one embodiment of the present disclosure. In this specification, the expression of the above terms is not necessarily for the same embodiment or example. Moreover, the specific features, structures or steps described can be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine the different embodiments or examples described in this specification and the features in different embodiments or examples, unless they are contradictory. In addition, it should be noted that in the present disclosure, the terms "first" and "second" are used for descriptive purposes only, and should not be understood as indicating or implying relative importance or order, nor should they be understood as implicitly indicating the number of technical features indicated.

[0044] In the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0045] In related technologies, various instant messaging software can be used to enable online communication between customers and customer service representatives, providing online customer service. However, these online customer service solutions cannot guarantee the confidentiality of customer information, posing information security risks. Therefore, this disclosure proposes a new session management solution for customer service scenarios, aiming to provide efficient and secure customer service and enhance customer trust and satisfaction with online customer service. The proposed session management solution will be described in detail below with reference to the accompanying figures.

[0046] Figure 1 An example scenario 100 is schematically shown to which the technical solutions provided by some embodiments of the present disclosure can be applied.

[0047] like Figure 1 As shown, application scenario 100 may include terminal device 110. Terminal device 110 may be, but is not limited to, a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, etc. A client for executing the session management method provided in some embodiments of the present disclosure may be deployed on terminal device 110, such as an independent application, a mini-program embedded in another application, or a browser for accessing a web application. This allows object 120 to initiate a session through the client and communicate with a matched target customer service representative.

[0048] The application scenario 100 may also include a server 130. The server 130 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal device 110 and the server 130 may be directly or indirectly connected via wired or wireless communication (for example, via the network 150), which is not limited in this application. For example, a server-side application for executing the session management method provided in some embodiments of the present disclosure may be deployed on the server 130 to provide online customer service.

[0049] In some embodiments, the application scenario 100 may further include a database 140. The terminal device 110 and / or the server 130 may be connected to the database 140 via the network 150 to exchange data with the database 140. Optionally, the database 140 may be an independent data storage device or device group, or may be a backend data storage device or device group related to other online services. For example, the database 140 may be used to store various types of data, such as data related to the alternative customer service mentioned in the embodiments below, and partial data related to the subject with the subject's consent.

[0050] In addition, in the present disclosure, the network 150 can be a wired network connected via cables, optical fibers, etc., or a wireless network such as 2G, 3G, 4G, 5G, Wi-Fi, Bluetooth, ZigBee, Li-Fi, etc., or an internal connection line of one or several devices, etc.

[0051] Figure 2 The example flow chart of the session management method 200 for customer service scenarios according to some embodiments of the present disclosure is schematically shown. Figure 1 Alternatively, the client may be an independent application client for providing customer service, or may be another type of client integrated with customer service functions, such as a client for a communication application, a client for a shopping application, etc. Figure 2 As shown, method 200 may include steps 210 to 250 .

[0052] In step 210, in response to receiving the object's session initiation operation, a session initiation request may be sent to the server. The session initiation request is used to initiate the current conversation and includes the object identifier of the object using the client. For example, a session initiation option, such as a button or menu option for initiating a conversation with customer service, may be presented in various interfaces (e.g., a communication or contact interface of a communication application, a browsing or add-to-purchase interface of a shopping application, etc.), and the object's session initiation operation may be received via the session initiation option. Alternatively, the object's session initiation operation may be received via a dialog box, such as receiving an instruction or statement from the object for initiating a conversation with customer service. Alternatively, the object's session initiation operation may be received via other appropriate means. After receiving the object's session initiation operation, a corresponding session initiation request may be sent to the server to initiate the current conversation between the object and customer service. The session initiation request may include the object identifier of the object currently using the client, such as the object's name, number, or other form of identifier.

[0053] At step 220, feedback information regarding the session initiation request is received from the server. The feedback information is configured to indicate that a target customer service has been determined, where the target customer service is a customer service that matches the object and is determined based on the object identifier. For example, after receiving the session initiation request, the server may determine a matching target customer service for the object based on the object identifier and send feedback information regarding the session initiation request to the client. The feedback information may indicate that a matching target customer service has been determined for the object and may also additionally indicate other information, such as information indicating that a session connection has been established between the object's client and the target customer service, and key information for the current session.

[0054] In step 230, in response to receiving the subject's session input operation, session information is sent to the server. The session information includes the session content entered by the subject through the session input operation. For example, after receiving feedback information from the server regarding the session initiation request, a session interface with the target customer service representative can be presented. The subject's session input operation can be received through this session interface, thereby receiving the subject's session content. Furthermore, based on the subject's session content, session information including the session content can be sent to the server. Optionally, the session input operation can be a text input operation, a voice input operation, a video input operation, or other input methods.

[0055] In step 240, in response to receiving a reply message from the server, the reply message is presented, wherein the reply message is generated by the target customer service based on the session information. For example, after receiving the session information, the target customer service can generate a reply to the session information on the server side and feed back the corresponding reply message to the client. For example, after receiving the reply message from the server, the client can present the reply message to the target. The presentation method of the reply message can be selected based on application requirements, for example, the reply message can be presented in text, voice, video, or other forms.

[0056] In step 250, in response to the subject exiting the current session, information indicating the end of the current session is sent to the server. Upon receiving the information indicating the end of the current session and determining that session deletion conditions are met, the server deletes at least a portion of the session records for the current session. The session deletion conditions include at least one of the following: the current session involves private data, the current session falls within a private session scenario, or a deletion instruction from the subject for at least a portion of the session records for the current session is received. Alternatively, the subject's exit from the current session can be determined by one or a combination of the following: the subject leaving the session interface for the current session, the subject remaining away from the session interface for a preset period of time, the subject logging out of the client, the subject clicking or otherwise selecting a button or option to exit the current session, or the subject inputting content indicating the end of the current session. For example, after determining that the subject has exited the current session, information may be sent to the server indicating the end of the current session. Upon receiving this information, the server may determine that the current session has ended. In this case, if the session deletion conditions are met, the server deletes at least a portion of the session records for the current session. Optionally, the operation of deleting at least a portion of the session record of the current session may be performed immediately after determining that the current session ends, or may be performed after a preset time period after determining that the current session ends.

[0057] For example, during the current session, the server can automatically detect whether sensitive or private data is involved, such as the name, ID number, ID photo, account number and password of the subject or other subjects of the current session. When sensitive or private data is detected, the portion of the session record of the current session related to such data can be automatically deleted. Alternatively, the server can automatically detect whether the scenario corresponding to the current session is a private session scenario, such as a healthcare consultation scenario, a legal affairs consultation scenario, a financial business consultation scenario, or a reporting scenario against other subjects or subjects' behaviors, which may involve a lot of private or sensitive information. When it is detected that the current session belongs to a private session scenario, all session records of the current session can be automatically deleted, or at least the portion involving private or sensitive information can be deleted. Alternatively, based on its own needs, when initiating the current session, during the current session, when exiting the current session, or after exiting the current session, the subject can actively choose to delete part or all of the session records of the current session through the client, or choose to adopt a session mode that deletes part or all of the session records. The subject's choice can be sent to the server, so that the server can delete all or part of the session records of the current session according to the subject's choice after the current session ends. Optionally, in an embodiment of automatic detection of private data or private conversation scenarios, different privacy protection levels can be pre-set, and the subject can select the desired privacy protection level to apply. For example, if the subject chooses to apply the lowest privacy protection level, the conversation record will be deleted only when highly sensitive or private data or scenarios are detected. If the subject chooses a higher privacy protection level, the conversation record will also be deleted when generally sensitive or private data or scenarios are detected. In addition, optionally, in an embodiment of automatic detection of private data or private conversation scenarios, when private data or private conversation scenarios are detected, a risk prompt can be presented to the subject through the client, and an option on whether to enable the conversation record deletion function can be presented alternatively or additionally, and then whether to delete the conversation record is determined based on the subject's selection. In addition, depending on the specific application scenario, other methods can also be used to determine whether the conversation record needs to be deleted and which part of the conversation record should be deleted.

[0058] Through method 200, in a scenario where a subject is engaging in a conversation with a target customer service representative, upon detecting that the current conversation involves private data or falls within a private conversation scenario, or based on a conversation record deletion instruction from the subject, at least a portion of the conversation record can be deleted after the subject exits the current conversation. This allows for the timely deletion of various relatively private information or information that the subject does not wish to retain after exiting the current conversation, preventing such information from being retained on the server. This significantly reduces the risk of such information being leaked and improves data security. Furthermore, this helps enhance the subject's trust in the customer service, allays or reduces their concerns about privacy leaks, ensures the safe and efficient delivery of customer service, and enhances the subject's satisfaction with the customer service.

[0059] More specifically, method 200 provides two modes for deleting conversation records: one for automatically detecting private data or private conversation scenarios, and the other for the user to proactively select a mode that deletes conversation records. This allows method 200 to be flexibly applied to a variety of customer service conversation scenarios requiring privacy protection. For example, it can be applied to customer service areas such as healthcare, finance, and law, where a high level of private or sensitive information may be involved. By leveraging method 200 to provide customer service, customers can confidently communicate with customer service without worrying about the leakage or misuse of their conversation records. Furthermore, method 200 can also be applied to scenarios requiring temporary needs. For example, if a customer wants to consult a specific question but does not wish for the consultation record to be permanently stored in the system, they can proactively select a mode that deletes some or all conversation records after the conversation ends, thereby satisfying their need to not retain conversation records. In summary, method 200 provides an effective means of protecting user privacy and data security, and is applicable to scenarios requiring user privacy protection and various temporary needs.

[0060] In some embodiments, the method 200 may further include: in response to the subject exiting the current session and satisfying a session deletion condition, deleting at least a portion of the session record for the current session. Similarly, the session deletion condition may include at least one of the following: the current session involves private data, the current session falls within a private session scenario, or a deletion instruction from the subject for at least a portion of the session record for the current session has been received. The determination of whether the subject has exited the current session and whether the session deletion condition has been satisfied are described in detail in step 250 and will not be repeated here. Optionally, the portion of the session record deleted by the client may be identical to or partially different from the portion deleted by the server, depending on the subject's settings. Optionally, the deletion of at least a portion of the session record for the current session may be performed immediately after the subject exits the current session, or may be performed after a preset period of time after the subject exits the current session. By deleting at least a portion of the session record for the current session on the client after the subject exits the current session, no sensitive data is retained on both the server and the client, thereby better protecting data security and preventing the leakage or misuse of private or sensitive data.

[0061] In some embodiments, to enable rapid recovery of historical session records and facilitate review of historical session records by the subject, the subject may be allowed to choose to store the session records locally, such as in local memory. For example, in response to the subject selecting to store at least a portion of the session record locally, a password setting control may be presented. The password setting control may receive a preset password entered by the subject, and the preset password may be applied to at least a portion of the session record. This allows the subject to view or otherwise manipulate the at least portion of the session record by setting a password. For example, when the subject logs out of an account or changes devices, the subject may restore the session record to the new account or device by uploading the locally stored session record. In this embodiment, the method 200 may further include: in response to receiving a processing operation for at least a portion of the locally stored session record, presenting a password verification control; receiving a verification password entered by the subject via the password verification control; and in response to the verification password matching the preset password, processing the at least portion of the session record according to the processing operation, wherein the preset password is a password pre-set by the subject to protect at least a portion of the locally stored session record. For example, the processing operations for at least a portion of the locally stored session record may include viewing, uploading, backing up, copying, and other operations. For example, if you want to upload part or all of a session log, you can use a password verification control to receive a verification password from the user before uploading. Only when the verification password matches a preset password will the session log be uploaded. For example, to ensure data security, a secure protocol such as SSL can be used to upload the session log to a cloud server.

[0062] In some embodiments, the feedback information in step 220 may include customer service characteristic data corresponding to the target customer service representative, and method 200 may further include presenting a virtual customer service avatar corresponding to the target customer service representative based on the customer service characteristic data. Exemplarily, the customer service characteristic data may be any form of characteristic data corresponding to the target customer service representative, such as characteristic data of an object matching the target customer service representative, characteristic data determined based on one or more of the object's behavior or attributes, or characteristic data describing a customer service avatar of the target customer service representative that matches the subject's preferences. This allows the client to present a virtual customer service avatar of the target customer service representative based on the customer service characteristic data after receiving it. Optionally, depending on specific application requirements, the virtual customer service avatar may be a two-dimensional or three-dimensional avatar. Furthermore, the virtual customer service avatar may also include settings for the customer service representative's movements, voice, and other characteristics. For example, different voice timbre, pitch, and speaking speed may be matched to the subject's preferences, or different movement amplitudes, movement types, and movement frequencies may be matched to the subject's preferences. For example, if the subject's behavior determines they have a lively and cheerful personality, they might prefer a more lively voice and larger movements; if their attributes determine they are older, they might prefer a slower speech rate and less frequent movements, and so on. Thus, based on the different preferences reflected in the subject's characteristic data, personalized virtual customer service avatars can be customized for each subject, helping to enhance the subject's conversation experience with the customer service and increase their satisfaction with the conversation.

[0063] In some embodiments, in order to further prevent information leakage and further ensure the security of conversation data between the subject and customer service, the risk of privacy data leakage can be reduced by detecting screenshots and blurring the screenshot images. In such embodiments, method 200 may also include: in response to detecting a screenshot operation for the current session, presenting a privacy protection option; in response to receiving a confirmation operation for the privacy protection option, blurring the privacy data in the screenshot image corresponding to the screenshot operation. Optionally, it is possible to determine whether the subject has implemented or completed the screenshot operation by monitoring the screenshot action, detecting the screenshot image, etc. For example, a service such as a custom AccessibilityService can be created, wherein the AccessibilityService is a service for helping the subject to better use Android devices, which can receive some events issued by the system, which can, for example, represent a series of state changes in the client's interface. Further illustratively, when a screenshot operation occurs, AccessibilityService may receive an event such as TYPE_WINDOW_STATE_CHANGED. Subsequently, by checking whether a screenshot image has been added to the system photo album, it can be confirmed whether the screenshot operation has been completed. Then, the screenshot image can be processed in a method such as onScreenshotTaken according to the privacy protection requirements of the object. For example, after detecting the screenshot operation, a privacy protection option can be presented, such as presenting a dialog box. In the dialog box, the object can be prompted that there is a risk of data leakage in the current screenshot operation, and the object can choose whether to perform privacy protection processing on the screenshot image. The privacy protection processing may include blurring the privacy data and other related processing, such as adding a watermark. If the object confirms to perform privacy protection processing on the privacy data in the screenshot image, the screenshot image corresponding to this screenshot operation can be obtained from the system photo album, and the privacy data in the screenshot image can be blurred, such as mosaic processing.

[0064] In some embodiments, the private data in the screenshot image corresponding to the screenshot operation can be obfuscated by: identifying a target keyword in the screenshot image, where the target keyword is a pre-set keyword related to the private data; determining the private data region based on the position of the target keyword in the screenshot image; and performing the obfuscation process on the private data region in the screenshot image. For example, text in the screenshot image can be extracted using OCR (Optical Character Recognition) technology, and keyword triggering rules can be pre-set so that when the extracted text contains the target keyword, the subsequent obfuscation process is triggered. Alternatively, the OCR text extraction process can be implemented using a pre-trained convolutional neural network (CNN). Optionally, the target keyword can be a keyword manually set in advance or automatically determined during the application operation, and can include, for example, an order number, password, amount, or other keywords that may involve private data. For example, after identifying the target keyword, the identified target keyword information can be matched with the original screenshot image to determine the position of the target keyword in the screenshot image, thereby determining the corresponding private data region. For example, the private data area can be an area within a preset range near the target keyword, or it can be an area within a preset range behind or below the target keyword, or, in the case of slot filling, the subject can fill in the corresponding data in the slot according to the customer service requirements, such as the order number, password, amount, etc., in which case such a slot can be regarded as a private data area. In addition, according to specific application requirements, other methods can also be used to determine the private data area based on the location of the target keyword. In this way, with the help of the triggering of the target keyword, the private data area can be quickly and accurately determined without over-analyzing the session records, so as to blur the private data.

[0065] For example, the private data region in the screenshot can be blurred by extracting it, filtering the extracted private data region to obtain a blurred private data region, and merging the blurred private data region with the screenshot to obtain a screenshot with the blurred private data. For example, a Gaussian blur process can be performed on the extracted private data region by applying a Gaussian filter to the region, blurring the private data region and blurring the private data region, thereby blurring the private data. Finally, the blurred private data region is merged with the screenshot to obtain a screenshot with the blurred private data. This blurred screenshot can effectively prevent the leakage of private data while meeting the needs of object storage or sharing session records.

[0066] In addition, in some embodiments, in order to intercept information leakage problems caused by screenshots, photos, etc., and to ensure the data security of the session interface, watermarks, etc. can also be introduced to avoid malicious screenshots or photos, and provide traceability when relevant information leaks. In this embodiment, method 200 may also include: presenting a watermark in the session interface for presenting the current session, wherein the watermark includes at least one of the following: object identifier, timestamp, session identifier. According to specific application requirements, a watermark in a unique and not easily ignored form can be pre-designed, and information related to the object or the current session can be retrieved to generate the watermark, such as one or more of the object account, timestamp, session number, etc., which can be used to uniquely identify the object or the current session. Optionally, a watermark can be added to the session interface of the current session using CSS (Cascading Style Sheets) style or Canvas.

[0067] In some embodiments, to ensure the security of the conversational content between the subject and the target customer service during transmission, the conversational content can be encrypted using end-to-end encryption technology to ensure that only the two parties in the current conversation (i.e., the subject using the client and the target customer service) can decrypt and view the conversational content. Specifically, after receiving a conversational input operation from the subject, the content entered by the subject through the conversational input operation can be encrypted using a first key to obtain conversational information, and the encrypted conversational information can be sent to the server. Furthermore, after receiving a reply from the server, the reply can be decrypted using a second key and the decrypted reply presented. The first and second keys can be pre-set keys used for encrypted transmission of the current conversation. Optionally, the first and second keys can be different or the same. For example, the conversational content between the subject and the target customer service can be encrypted using AES (Advanced Encryption Standard), a symmetric encryption algorithm. When AES encryption is used, the first and second keys can be the same key. Optionally, the first and second keys can be automatically or randomly generated according to preset rules, or can be specified by the user. By leveraging the above encryption scheme, combined with instant messaging technology, a secure, real-time, and private communication environment can be created, which helps further enhance user satisfaction and trust in customer service.

[0068] For example, in some embodiments, the security of the session can be further enhanced based on digital certificates. For example, the server can generate its own public key and private key, apply for a certificate from a certification authority, and submit the public key to the certification authority. The certification authority can create and issue a certificate for the server. Subsequently, when a session connection is established between the target customer service and the client of the object, the server can send a certificate to the client, and the client can verify the identity of the server through the certificate. After the verification is passed, the client can generate a random symmetric key, encrypt the symmetric key with the server's public key, and send it to the server, so that subsequent session content can be encrypted with the symmetric key. In addition, the identity of the object can also be verified based on a personal digital certificate to prevent identity impersonation, thereby further enhancing session security. For example, similar to the server, the object can generate its own key pair (including public and private keys), register its own public key with the certification authority, and obtain a certificate, which can be used to verify the identity of the object when initiating a subsequent session.

[0069] Figure 3 The example flow chart of the session management method 300 for customer service scenarios according to some embodiments of the present disclosure is schematically shown. Figure 1 The server 130 shown is implemented. Figure 3 As shown, method 300 may include steps 310 to 360 .

[0070] At step 310, a session initiation request may be received from the client. The session initiation request may be used to initiate the current session and may include the object identifier of the object using the client. As described with respect to step 210, upon receiving the session initiation operation for the object, the client may send the session initiation request to the server. The session initiation request may include the object identifier of the object currently using the client, such as the object's name, number, or other identifier.

[0071] In step 320, a target customer service representative that matches the object can be determined based on the object identifier. For example, based on the object identifier, a customer service representative that the object has previously used and rated well can be obtained as the target customer service representative. Alternatively, for example, based on the object identifier, a customer service representative that has been used and rated well by an object or object group similar to the object can be determined as the target customer service representative. Alternatively, for example, based on the object identifier, historical conversations between the object and the customer service representative can be obtained, such as partial conversation content of historical conversations, satisfaction levels of historical conversations, conversation topics of historical conversations, etc. Alternatively, other behaviors or behavioral preferences of the object can be obtained, such as payment behaviors, shopping history, product category preferences, product style preferences, music style preferences, web browsing preferences, etc. Alternatively, attribute information of the object can be obtained, such as age, occupation, gender, etc. Furthermore, based on this obtained information, a target customer service representative that matches the object can be determined. In addition, depending on specific application requirements, a matching target customer service representative can also be determined based on other information of the object. Exemplarily, the above information can be obtained based on the consent of the object. For example, the object can be asked whether it agrees to collect or obtain its various behaviors or behavioral preferences. For example, with the consent of the object, partial session records of historical sessions that do not involve privacy data can be stored. For example, only the public attribute information of the object can be obtained, or its attribute information can be obtained based on the consent of the object, and so on.

[0072] In step 330, feedback information regarding the session initiation request may be sent to the client. The feedback information is configured to indicate that a target agent has been identified. As described with respect to step 220, in addition to indicating that a matching target agent has been identified for the subject, the feedback information may also include other information, such as information indicating that a session connection has been established between the subject's client and the target agent, and key information for the current session. Optionally, the feedback information may also include agent characteristic data corresponding to the target agent, so that the client can present a virtual agent avatar corresponding to the target agent based on the agent characteristic data.

[0073] In step 340, session information may be received from the client, the session information including the session content input by the subject through the client. As described with respect to step 230, the client may send session information to the server based on the subject's session input operation, the session information including the session content input by the subject through the session input operation.

[0074] At step 350, a response message can be generated and sent to the client via the target customer service based on the received session information. For example, upon receiving the session information, the server can generate a response to the session information via the target customer service and send the corresponding response message back to the client. For example, the target customer service can be a trained and debugged machine learning model that can input the received session information into the model and generate a response message based on the model output.

[0075] In step 360, in response to the current session ending and satisfying a session deletion condition, at least a portion of the session record for the current session may be deleted. The session deletion condition includes at least one of the following: the current session involves private data, the current session belongs to a private session scenario, and a deletion instruction has been received from the subject for at least a portion of the session record for the current session. As described with respect to step 250, when the subject exits the current session, the client may send information indicating the end of the current session to the server. Upon receiving this information, the server may determine that the current session has ended. How to determine whether the subject has exited the current session and how to delete at least a portion of the session record based on the session deletion condition have been described in detail in step 250 and will not be repeated here.

[0076] Method 300 may have the same embodiments as method 200 and may achieve the same technical effects. For the sake of brevity, the description is not repeated here.

[0077] In some embodiments, step 320 may include: obtaining feature data corresponding to the object based on the object identifier, the feature data representing at least one of the object's behavior and object attributes, and determining a target customer service representative matching the object based on the feature data. For example, with the object's consent, at least one of various behavioral data and attribute data associated with the object may be obtained based on the object identifier, such as the object's historical conversations with the customer service representative, the object's other behaviors or behavioral preferences such as shopping, listening to music, and web browsing, and the object's attribute information such as age, occupation, and gender. After obtaining the above data, feature data, such as feature vectors (e.g., embeddings), may be obtained by encoding the data according to preset rules, processing the data using a pre-trained encoding model, or processing the data in other ways. Alternatively, the server may periodically update the object's feature data based on the object's related behavioral data, attribute data, etc., so that the corresponding feature data can be read directly based on the object identifier. Alternatively, the object's feature data may be read directly from another server or database based on the object identifier.

[0078] Exemplarily, for the behavior of an object, such as the conversation records of the historical conversations between the object and the customer service, natural language processing technology can be used to extract features therein, such as word frequency, part of speech, sentiment analysis, etc. For the attributes of the object, they can be encoded according to preset rules. Exemplarily, the features extracted based on the behavior of the object and the encoding generated based on the attributes of the object can be combined, and optionally processed in a specific way to constitute the feature data of the object. Alternatively, the feature data of the object can also be generated in other ways, and the present disclosure does not impose specific restrictions on this. Exemplarily, the feature data generated based on one or more information in the behavior and attributes of the object can be classified and labeled, such as labeling the object's emotions, personality, interests, etc., which can be implemented, for example, with the help of a preset algorithm or a pre-trained model. Exemplarily, the target customer service that matches the object can be determined based on the above-mentioned feature data or the classified and labeled information. In some embodiments of the present disclosure, the customer service can be a customer service robot, which can be implemented by a pre-trained machine learning model. Exemplarily, a customer service robot can receive startup data, where the startup data can be used to set the customer service robot's personality or conversational style through pre-formatted instructions, sample corpus, and the like, such as language style (e.g., cute, steady, concise, detailed, etc.), topic preferences, whether to use dialects, frequency of emoticons, etc., so that the customer service robot can be customized to a target customer service agent that matches the subject based on this startup data. Exemplarily, after obtaining the subject's feature data or classification annotation information, corresponding startup data can be generated or acquired based on the feature data or classification annotation information, and the generated or acquired startup data can be provided to the customer service robot to obtain the determined target customer service agent. Alternatively, the target customer service agent can be determined through other means. For example, input data for controlling the customer service agent's conversational style can be determined based on the subject's feature data, and this input data can be input into a customer service model alone or together with conversation information (or conversation content) so that the customer service model can generate a response in a desired manner. The process of determining the input data for controlling the customer service agent's conversational style can be considered the process of determining the target customer service agent. Alternatively, the target customer service agent can be determined based on the subject's feature data in other suitable ways. Optionally, during subsequent conversations, the target customer service may further adjust his or her personality or conversation style based on the conversation information sent by the subject to better match the subject's personality or conversation style, or in other words, better match the subject's preferred personality or conversation style, so as to provide the subject with a better customer service experience.

[0079] By determining the target customer service that matches the object based on the object's characteristic data, and determining personalized exclusive customer service that is similar to the object's personality or conversation style, this helps optimize the fluency of the conversation between the object and the customer service, improve conversation efficiency, and enhance the conversation experience.

[0080] In some embodiments, multiple candidate customer service representatives can be determined based on data from training or use of a customer service model. Each candidate customer service representative can have feature data, which can be feature data of an object that matches the candidate customer service representative. Here, the feature data of the object that matches the candidate customer service representative can be feature data corresponding to an actual object, a collection of feature data of a group of objects, or feature data corresponding to a class of objects. Optionally, depending on the customer service implementation, different candidate customer service representatives can be defined by one or more of different model structures, different model parameters, different model startup data, different input data for controlling the customer service representative's conversational style, and so on. In such an embodiment, the similarity between the feature data of the object and the feature data of each candidate customer service representative in the multiple candidate customer service representatives can be determined, and then the target customer service representative that matches the object can be determined based on the candidate customer service representative with the highest similarity. Exemplarily, the feature data can be a feature vector, or data that can be converted into vector form. The similarity between different vectors can be measured using, for example, cosine similarity, Euclidean distance, Manhattan distance, or the like. Alternatively, the similarity between different feature data or vectors can be determined using a pre-trained similarity model. By determining the target customer service representative from multiple candidate customer service representatives based on the similarity of feature data, we can leverage the advantages of big data to more quickly and accurately determine matching target customer service representatives for the target, thereby improving the efficiency and quality of determining the target customer service representative, and thus helping to improve the target customer service conversation experience.

[0081] In some embodiments, to better match the subject's conversational style or preferred conversational style, when determining a target customer service representative based on the most similar candidate customer service representatives, the target customer service representative can be determined by adjusting the conversational style of the most similar candidate customer service representative based on feature data to match the subject's preferred conversational style as reflected in the feature data. Based on the similarity of the feature data, an object or class of objects that is most similar to the current subject can be determined among the objects matching the candidate customer service representatives. The current subject and the determined object or class of objects generally have the same or similar conversational style or conversational style preferences. However, due to the uniqueness of different subjects, the determined target customer service representative often cannot fully match all of the subject's needs and preferences. By adjusting the conversational style of the most similar candidate customer service representative based on feature data, it is possible to better match the subject's conversational style preferences, more comprehensively achieve personalized customization of the target customer service representative, and thus further enhance the subject's customer service conversation experience. For example, the conversational style of the most similar candidate customer service representative can be adjusted based on the differences in behavioral or attribute characteristics between the current subject and the object matching the most similar candidate customer service representative. For example, assuming there is an age difference between the two and the current subject is older, the frequency of using emoticons can be reduced or eliminated, and the speaking speed can be reduced during the conversation. assuming there are regional differences between the two, the language habits can be adjusted based on the region of the current subject to match the dialect and idioms of their region. assuming there are differences in punctuation usage between the two, such as the current subject tends to use more different punctuation to express different tones, more different punctuation can be used during the conversation to reflect different tones. Optionally, depending on the different implementation methods of customer service, the above-mentioned personalized adjustments can be made in corresponding ways, such as generating supplementary startup data, adjusting startup data, adjusting input data used to control the customer service's conversational style, etc., to adjust the conversational style of the candidate customer service with the highest similarity.

[0082] In some embodiments, the method 300 may further include: obtaining a subject's conversation evaluation of the current conversation; and, in response to the conversation evaluation satisfying a preset evaluation threshold, determining a new candidate or updating the corresponding candidate based on the target customer service. Exemplarily, during or after the current conversation, a conversation evaluation prompt may be presented to the subject via the client. For example, a pop-up window, dialog box, or conversation content may be presented for obtaining a review of the current conversation. Alternatively, the subject may be allowed to obtain the subject's conversation evaluation of the current conversation by inputting or selecting a rating score, rating, or content. Accordingly, the preset rating threshold may optionally include a rating score threshold, a rating threshold, a threshold for the degree of positive reviews reflected in the review content, or the inclusion of preset positive keywords. Exemplarily, data related to the target customer service may be stored, such as feature data, behavioral or attribute characteristics of subjects matching the target customer service, customer service feature data of the target customer service, and data or information used to define the target customer service, to obtain a new candidate. Alternatively, the candidate may be updated by updating the data related to the target customer service, based on the data related to the target customer service. By adding or updating alternative customer service representatives based on conversation evaluations, it is helpful to improve the richness of alternative customer service representatives, as well as the matching degree between alternative customer service representatives and different target groups and the degree of generalization in the corresponding target groups. This helps to improve the efficiency of determining target customer service representatives based on alternative customer service representatives, and helps to improve the matching degree between the determined target customer service representatives and the target, thereby optimizing the target customer service conversation experience.

[0083] In some embodiments, after deleting at least a portion of the session history for the current session, the subject can be notified in real time to further enhance the subject's trust and satisfaction with data security. Specifically, the method 200 may further include the following steps: sending a session deletion message to the client, indicating that at least a portion of the session history for the current session has been deleted.

[0084] In some embodiments, as described in the above embodiments, in order to ensure the security of the conversation interaction content between the object and the target customer service during transmission, the conversation interaction content can be encrypted using end-to-end encryption technology to ensure that only the two parties in the current conversation (i.e., the object using the client and the target customer service) can decrypt and view the conversation content. In this embodiment, the aforementioned step 350 may include: decrypting the received conversation information according to the first key; generating reply content through the target customer service based on the decrypted conversation information; encrypting the generated reply content according to the second key to obtain reply information; and sending the reply information to the client, wherein the first key and the second key are pre-set keys for encrypted transmission of the current session. Optionally, depending on the encryption method, the first key and the second key used by the server can be the same as or different from the first key and the second key used by the client. In addition, the embodiments and corresponding technical effects of the use and acquisition of keys have been described in detail above and will not be repeated here for the sake of brevity.

[0085] In some embodiments, to provide a better conversation experience and higher conversation efficiency, seamless contextual customer service support can be implemented by providing a backtracking function for historical conversation records, further improving customer service conversation efficiency and optimizing the customer service conversation experience. For example, step 350 may include: in response to the conversation information containing information related to the historical conversation, obtaining at least a portion of the conversation record of the historical conversation; and generating and sending a reply message to the client via the target customer service representative based on at least the portion of the conversation record and the conversation information. For example, in the current conversation, the conversation content entered by the client through the client may refer to information from a previous historical conversation, such as questions, topics, or conclusions mentioned in a previous conversation or a conversation on a certain date. After receiving the conversation information containing such conversation content, the target customer service representative may retrieve the relevant conversation record of the corresponding historical conversation based on identifiers related to the historical conversation in the conversation content, such as the time of the conversation mentioned by the client, the previous conversation or similar expressions, and keywords related to the questions, topics, or conclusions mentioned in the conversation, so as to generate a reply message based on both the retrieved conversation record and the currently received conversation information.

[0086] As will be appreciated by those skilled in the art, although Figure 2 、 3 and about Figure 2 、 3The description of the method provided by the present disclosure describes the steps of the method in a particular order, but this does not require or imply that the steps must be performed in this particular order unless the context clearly indicates otherwise. Additionally or alternatively, multiple steps can be combined into a single step and / or a single step can be broken down into multiple steps. In addition, other method steps can be inserted between steps. Inserted steps can represent improvements to the method described herein or can be unrelated to the method. In addition, a given step may not be fully completed before the next step begins.

[0087] To further facilitate understanding, Figure 4 Schematically illustrates an example interaction diagram of a session management method for a customer service scenario according to some embodiments of the present disclosure. Figure 4 As shown, the client can obtain various types of object behaviors 401, such as conversation behaviors, purchase behaviors, browsing behaviors, etc., and the client can upload behavior data 402 to the server based on the obtained object behaviors. The server can extract the feature data of the object based on the uploaded behavior data through the various methods described in the previous embodiments, and determine the target customer service 403 based on the feature data. In addition, although Figure 4(Not shown) The server may also use object attribute data obtained based on the object identifier or attribute data provided by the object through the client. This attribute data can also be used to determine the object's characteristic data and to identify the target customer service representative. The server may return a result 402 to the client, informing the client that a target customer service representative matching the object has been determined, and optionally send customer service characteristic data related to the target customer service representative to the client so that the client can present the target customer service representative's avatar 403 to the object. Optionally, some or all of steps 401 through 405 may be performed after the object initiates the current session, or some of these steps may be performed periodically or on demand. For example, step 401 may occur based on the object's operation, step 402 may be performed periodically or when the object performs an action, step 403 may be performed periodically or when the object initiates the current session, and so on. During the current session, the object may enter session content 406 by performing a session input operation on the client, and the client may upload session information 407 to the server based on the entered session content. At the server, the target customer service can be instructed to generate a reply 408. Specifically, the target customer service can be instructed to generate reply information based on the conversation content in the received conversation information. The server can return a result 409 to the client to feedback the reply information for the aforementioned conversation information, and the client can present 410 the reply information to the object in an appropriate presentation manner. Optionally, in a conversation, steps 406 to 410 can be performed once or multiple times. When the client determines that the object has exited the current conversation, it can send a session end message 411 to the server to indicate that the current conversation has ended. After receiving the message, the server can determine that the current conversation has ended. At this time, if the session deletion condition is met, the operation of deleting the session record 412 is performed to delete at least part of the conversation record of the current conversation. After the conversation record is deleted, the server can return a result to the client to inform it that at least part of the conversation record of the current conversation has been deleted. In addition, during or after the current conversation, when the client detects the screenshot operation 414 of the object, it can take the screenshot protection 415 operation, such as blurring the privacy data of the screenshot image based on the confirmation operation of the object. About Figure 4 The specific implementation of each step in the above has been described in detail, and for the sake of brevity, it will not be repeated here.

[0088] Figure 5 、 Figure 6 Schematic diagrams of example architectures of a client and a server for executing the session management method according to some embodiments of the present disclosure are shown respectively. Figure 5 、 Figure 6As shown, the client and server of the above session management method can be implemented by relying on the client 500 and server 600 of the instant messaging application, so as to ensure the immediacy of the session between the object and the customer service by relying on the instant messaging service.

[0089] like Figure 5 As shown, the instant messaging application client 500 may include a customer service interface module 510, which may be configured to execute various embodiments of the method 200 described above to provide customer service conversation services. Exemplarily, the customer service interface module 510 may include one or more of an identity recognition module 511, a privacy processing module 512, a customer service generation module 513, and an object conversation module 514. The identity recognition module 511 may be configured to interact with corresponding modules on the server to perform object identity authentication. Exemplarily, this module may be implemented based on a framework such as OpenID. The privacy processing module 512 may be configured to perform operations related to object privacy protection, such as deleting portions of conversation records after a conversation ends in response to meeting a session deletion condition, blurring a screenshot in response to an object screenshot operation, and so on. The customer service generation module 513 may be configured to generate a virtual customer service avatar of a target customer service representative based on customer service characteristic data provided by the server. The object conversation module 514 may be configured to interact with corresponding modules on the server to provide conversation services. Alternatively or additionally, the client 500 may further include other modules according to specific application requirements.

[0090] like Figure 6As shown, the instant messaging application server 600 may include an encryption service module 610, an identity authentication service module 620, an instant messaging module 630, and a scheduled task module 640. The encryption service module 610 may be configured to provide encryption and decryption of the conversation content between the subject and the customer service. For example, it may include a message encryption module 611 and a message decryption module 612. These two modules may perform decryption operations on the client's session information and encryption operations on the customer service generated replies using a pre-set or real-time assigned key. The identity authentication service module 620 may be configured to provide services related to identity authentication. For example, it may include an object identity verification module 621 and a target customer service generation module 622. The object identity verification module 621 may, for example, interact with the identity recognition module 511 in the client 500 to implement object identity verification. The target customer service generation module 622 may be configured to determine a target customer service that matches the subject based on the subject's object identifier. The instant messaging module 630 may be configured to provide instant messaging services. The scheduled task module 640 can be configured to provide time-related services, which may include, for example, a timestamp module 641 and an immediate destruction module 642. The timestamp module 641 can be used to pull out timestamps, and the pulled timestamps can be used to generate watermark information presented in the session interface, for example. The immediate destruction module 642 can be used to destroy session records (i.e., delete session records), for example, destroying at least a portion of the session records after the object exits the session or after a preset time after exiting the session. In addition, exemplarily, the server 600 may also include a storage module for storing various types of session records and other data, such as a picture storage module 650, a text storage module 660, and a video storage module 670. Alternatively or additionally, the server 600 may also include other modules according to specific application requirements.

[0091] Figure 7 The following schematically illustrates an example block diagram of a session management apparatus 700 for customer service scenarios according to some embodiments of the present disclosure. As shown in the figure, the apparatus 700 may include a session initiating module 710, a receiving module 720, a first sending module 730, a presenting module 740, and a second sending module 750.

[0092] Specifically, the session initiation module 710 may be configured to: in response to receiving a session initiation operation of an object, send a session initiation request to the server, the session initiation request is used to initiate the current conversation and includes the object identifier of the object using the client; the receiving module 720 may be configured to: receive feedback information from the server for the session initiation request, the feedback information is configured to indicate that the target customer service has been determined, wherein the target customer service is the customer service that matches the object determined based on the object identifier; the first sending module 730 may be configured to: in response to receiving a session input operation of the object, send session information to the server, the session information includes the session content entered by the object through the session input operation The presentation module 740 may be configured to: in response to receiving reply information from the server, present the reply information, wherein the reply information is generated by the target customer service based on the session information; the second sending module 750 may be configured to: in response to the object exiting the current session, send information indicating the end of the current session to the server, so that when the server receives the information indicating the end of the current session and determines that the session deletion condition is met, it deletes at least a part of the session record of the current session, and the session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction from the object for at least a part of the session record of the current session is received.

[0093] Figure 8 The following schematically illustrates an example block diagram of a conversation management apparatus 800 for customer service scenarios according to some embodiments of the present disclosure. As shown, the apparatus 800 may include a first receiving module 810, a determining module 820, a sending module 830, a second receiving module 840, a reply module 850, and a deletion module 860.

[0094] Specifically, the first receiving module 810 can be configured to: receive a session initiation request from the client, the session initiation request is used to initiate the current conversation and includes an object identifier of an object using the client; the determination module 820 can be configured to: determine the target customer service that matches the object based on the object identifier; the sending module 830 can be configured to: send feedback information regarding the session initiation request to the client, and the feedback information is configured to indicate that the target customer service has been determined; the second receiving module 840 can be configured to: receive session information from the client, and the session information includes the session content input by the object through the client; the reply module 850 can be configured to: generate and send reply information to the client through the target customer service based on the received session information; the deletion module 860 can be configured to: delete at least a part of the session record of the current session in response to the end of the current session and the satisfaction of the session deletion condition, the session deletion condition including at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction is received from the object for at least a part of the session record of the current session.

[0095] It should be understood that the apparatuses 700 and 800 can be implemented in software, hardware, or a combination of software and hardware. Multiple different modules can be implemented in the same software or hardware structure, or one module can be implemented by multiple different software or hardware structures.

[0096] Furthermore, apparatus 700 can be used to implement method 200 described above, and apparatus 800 can be used to implement method 300 described above. The relevant details have been described in detail above and will not be repeated here for the sake of brevity. Apparatuses 700 and 800 can have the same features and advantages as those described with respect to methods 200 and 300.

[0097] Figure 9 Schematically illustrates an example block diagram of a computing device 900 according to some embodiments of the present disclosure. For example, it may represent Figure 1 The terminal device 110 or server 130 in.

[0098] As shown, the example computing device 900 includes a processing system 901, one or more computer-readable media 902, and one or more I / O interfaces (input / output interfaces) 903 that are communicatively coupled to each other. Although not shown, the computing device 900 may also include a system bus or other data and command transmission system that couples the various components to each other. The system bus may include any one or a combination of different bus structures, such as a memory bus or memory controller, a peripheral bus, a universal serial bus, and / or a processor or local bus utilizing any of a variety of bus architectures, or may also include control and data lines.

[0099] Processing system 901 represents the functionality of performing one or more operations using hardware. Thus, processing system 901 is illustrated as including hardware elements 904 that can be configured as processors, functional blocks, and the like. This can include implementing application-specific integrated circuits (ASICs) or other logic devices formed using one or more semiconductors in hardware. Hardware elements 904 are not limited by the materials from which they are formed or the processing mechanisms employed therein. For example, a processor can be comprised of (a plurality of) semiconductors and / or transistors (e.g., electronic integrated circuits (ICs)). In such a context, processor-executable instructions can be electronically executable instructions.

[0100] The computer-readable medium 902 is illustrated as including a memory / storage device 905. The memory / storage device 905 represents a memory / storage device associated with one or more computer-readable media. The memory / storage device 905 may include volatile storage media (such as random access memory (RAM)) and / or non-volatile storage media (such as read-only memory (ROM), flash memory, optical disk, magnetic disk, etc.). The memory / storage device 905 may include fixed media (e.g., RAM, ROM, fixed hard drive, etc.) and removable media (e.g., flash memory, removable hard drive, optical disk, etc.). Exemplarily, the memory / storage device 905 can be used to store the various behavioral data, attribute data, and session records associated with the objects mentioned in the above embodiments. The computer-readable medium 902 can be configured in various other ways as further described below.

[0101] One or more I / O interfaces 903 represent functionality that allows an object to enter commands and information into the computing device 900 and also allows information to be presented to the object and / or sent to other components or devices using various input / output devices. Examples of input devices include a keyboard, a cursor control device (e.g., a mouse), a microphone (e.g., for voice input), a scanner, a touch function (e.g., a capacitive or other sensor configured to detect physical touch), a camera (e.g., that can use visible or invisible wavelengths (such as infrared frequencies) to detect motion that does not involve touch as gestures), a network card, a receiver, and the like. Examples of output devices include a display device (e.g., a monitor or projector), a speaker, a printer, a tactile response device, a network card, a transmitter, and the like. Exemplarily, in the embodiments described above, the input devices can allow the object to perform various interactive operations, such as initiating a conversation, entering conversation content, and the like, and the output devices can allow the object to view various presented content, such as presented reply information to conversation information, a virtual customer service image, and the like.

[0102] The computing device 900 also includes a session management application 906. The session management application 906 may be stored as computer program instructions in the memory / storage device 905. The session management application 906 may be implemented in conjunction with the processing system 901 or the like regarding Figure 7 or Figure 8 The entire functions of each module of the session management device 700 or 800 are described.

[0103] Various techniques may be described herein in the general context of software, hardware, elements, or program modules. Generally, these modules include routines, programs, objects, elements, components, data structures, etc. that perform specific tasks or implement specific abstract data types. As used herein, the terms "module," "function," etc. generally refer to software, firmware, hardware, or a combination thereof. A feature of the techniques described herein is that they are platform-independent, meaning that these techniques can be implemented on a variety of computing platforms with a variety of processors.

[0104] An implementation of the described modules and techniques may be stored on or transmitted across some form of computer-readable media. Computer-readable media may include various media accessible by the computing device 900. By way of example and not limitation, computer-readable media may include "computer-readable storage media" and "computer-readable signal media."

[0105] As opposed to a mere signal transmission, carrier wave, or signal itself, "computer-readable storage media" refers to media and / or devices, and / or tangible storage devices, capable of persistently storing information. Thus, computer-readable storage media refers to non-signal-bearing media. Computer-readable storage media include hardware such as volatile and non-volatile, removable and non-removable media and / or storage devices implemented in a method or technology suitable for storing information (such as computer-executable instructions, data structures, program modules, logic elements / circuits, or other data). Examples of computer-readable storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVDs) or other optical storage devices, hard disks, cassettes, magnetic tape, magnetic disk storage devices or other magnetic storage devices, or other storage devices, tangible media, or articles of manufacture suitable for storing desired information and accessible by a computer.

[0106] "Computer-readable signal media" refers to signal-bearing media configured to transmit instructions to the hardware of computing device 900, such as via a network. Signal media typically embodies computer-executable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, data signal, or other transport mechanism. Signal media also includes any information transmission media. By way of example, and not limitation, signal media include wired media such as a wired network or direct connection, and wireless media such as acoustic, RF, infrared, and other wireless media.

[0107] As previously mentioned, hardware elements 904 and computer-readable medium 902 represent instructions, modules, programmable device logic and / or fixed device logic implemented in hardware form, which can be used to implement at least some aspects of the technology described herein in some embodiments. Hardware elements can include other implementations in integrated circuits or systems on a chip, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs) and silicon or components of other hardware devices. In this context, hardware elements can be used as processing equipment for executing program tasks defined by the instructions, modules and / or logic embodied by the hardware elements, and hardware devices for storing instructions for execution, such as the computer-readable storage media previously described.

[0108] The aforementioned combinations may also be used to implement the various techniques and modules described herein. Thus, software, hardware or program modules and other program modules may be implemented as one or more instructions and / or logic embodied on some form of computer-readable storage medium and / or by one or more hardware elements 904. The computing device 900 may be configured to implement specific instructions and / or functions corresponding to the software and / or hardware modules. Thus, for example, by using a computer-readable storage medium and / or hardware elements 904 of a processing system, a module may be implemented as a module executable by the computing device 900 as software, at least in part, in hardware. Instructions and / or functions may be executed / operable by, for example, one or more computing devices 900 and / or processing systems 901 to implement the techniques, modules, and examples described herein.

[0109] The techniques described herein may be supported by these various configurations of computing device 900 and are not limited to the specific examples of the techniques described herein.

[0110] It should be understood that, for the sake of clarity, embodiments of the present disclosure have been described with reference to different functional units. However, it will be apparent that, without departing from the present disclosure, the functionality of each functional unit can be implemented in a single unit, in multiple units, or as a part of other functional units. For example, the functionality described as being performed by a single unit can be performed by multiple different units. Therefore, reference to a specific functional unit is only considered as a reference to the appropriate unit for providing the described functionality, rather than indicating a strict logical or physical structure or organization. Therefore, the present disclosure can be implemented in a single unit, or can be physically and functionally distributed between different units and circuits.

[0111] The present disclosure provides a computer-readable storage medium having computer-executable instructions stored thereon. When the computer-executable instructions are executed, the above-mentioned session management method for customer service scenarios is implemented.

[0112] The present disclosure provides a computer program product or computer program, which includes computer-executable instructions stored in a computer-readable storage medium. A processor of a computing device reads the computer-executable instructions from the computer-readable storage medium and executes the computer-executable instructions, causing the computing device to implement the conversation management method for customer service scenarios provided in the various embodiments described above.

[0113] It is understood that in the specific embodiments of this disclosure, behavioral data, attribute data, etc. of an object are involved. When the embodiments described in this disclosure involving such data are applied to specific products or technologies, the permission or consent of the object must be obtained, and the collection, use, and processing of the relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0114] Variations on the disclosed embodiments will be understood and effected by those skilled in the art in practicing the claimed subject matter by studying the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

Claims

1. A conversation management method for a customer service scenario, comprising: In response to receiving the session initiation operation of the object, sending a session initiation request to the server, wherein the session initiation request is used to initiate the current conversation and includes an object identifier of the object using the client; receiving feedback information regarding the session initiation request from the server, wherein the feedback information is configured to indicate that a target customer service has been determined, wherein the target customer service is a customer service that matches the object and is determined based on the object identifier; In response to receiving the conversation input operation of the object, sending conversation information to the server, the conversation information including conversation content input by the object through the conversation input operation; In response to receiving reply information from the server, presenting the reply information, wherein the reply information is generated by the target customer service based on the session information; In response to the object exiting the current session, information indicating the end of the current session is sent to the server, so that when the server receives the information indicating the end of the current session and determines that a session deletion condition is met, the server deletes at least a portion of the session record of the current session, and the session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion indication from the object for at least a portion of the session record of the current session is received.

2. The method according to claim 1, wherein The feedback information includes customer service characteristic data corresponding to the target customer service, and wherein the method further includes: Based on the customer service characteristic data, a virtual customer service image corresponding to the target customer service is presented.

3. The method according to claim 1, further comprising: In response to detecting a screenshot operation for the current session, presenting a privacy protection option; In response to receiving a confirmation operation for the privacy protection option, a blurring process is performed on the private data in the screenshot image corresponding to the screenshot operation.

4. The method according to claim 3, wherein: The blurring of the private data in the screenshot image corresponding to the screenshot operation includes: identifying target keywords in the screenshot image, where the target keywords are pre-set keywords related to the privacy data; determining a private data area according to a position of the target keyword in the screenshot image; A blurring process is performed on the private data area in the screenshot image.

5. The method according to claim 4, wherein The performing blurring processing on the private data area in the screenshot image includes: extracting the private data area in the screenshot image; Filtering the extracted private data region to obtain a blurred private data region; The blurred private data area is merged with the screenshot image to obtain a screenshot image with blurred private data.

6. The method according to claim 1, further comprising: In response to the object exiting the current session and satisfying the session deletion condition, at least a portion of the session record of the current session is deleted.

7. The method according to claim 1, further comprising: In response to receiving a processing operation for at least a portion of the locally stored session record, presenting a password verification control; receiving a verification password input by the object through the password verification control; In response to the verification password matching a preset password, processing the at least part of the session record according to the processing operation, wherein the preset password is a password preset by the object for protecting at least part of the locally stored session record.

8. The method according to claim 1, wherein The sending of session information to the server comprises: encrypting the content input by the object through the session input operation according to the first secret key to obtain session information, and sending the session information to the server, and The presenting of the reply information includes: Decrypt the reply message according to the second secret key and present the decrypted reply message. The first key and the second key are pre-set keys used for encrypted transmission of the current session.

9. A conversation management method for a customer service scenario, comprising: receiving a session initiation request from a client, wherein the session initiation request is used to initiate a current conversation and includes an object identifier of an object using the client; Based on the object identifier, determining a target customer service that matches the object; Sending feedback information regarding the session initiation request to the client, wherein the feedback information is configured to indicate that the target customer service has been determined; receiving session information from the client, wherein the session information includes session content input by the subject through the client; Based on the received session information, generating and sending a reply message to the client through the target customer service; In response to the current session ending and satisfying a session deletion condition, at least a portion of the session record of the current session is deleted. The session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion indication is received from the object for at least a portion of the session record of the current session.

10. The method according to claim 9, wherein: Determining a target customer service matching the object based on the object identifier includes: Based on the object identifier, acquiring feature data corresponding to the object, the feature data representing at least one of an object behavior and an object attribute of the object; Based on the feature data, a target customer service representative matching the object is determined.

11. The method according to claim 10, wherein: Determining a target customer service representative matching the object based on the feature data includes: Determining a similarity between the feature data and feature data of each candidate customer service representative among a plurality of candidate customer service representatives, wherein the feature data of each candidate customer service representative is feature data of an object that matches the candidate customer service representative; The target customer service is determined based on the candidate customer service with the highest similarity.

12. The method according to claim 11, wherein The determining of the target customer service representative based on the candidate customer service representative with the highest similarity includes: The target customer service representative is determined by adjusting the conversation style of the candidate customer service representative with the highest similarity based on the feature data to match the conversation style preferred by the subject as reflected by the feature data.

13. The method according to claim 11 or 12, further comprising: Obtaining a session evaluation of the object on the current session; In response to the conversation evaluation meeting a preset evaluation threshold, a new candidate customer service is determined based on the target customer service or a corresponding candidate customer service is updated.

14. The method according to claim 9, wherein The generating, by the target customer service, and sending a reply message to the client based on the received session information includes: decrypting the received session information according to the first secret key; Based on the decrypted session information, the target customer service generates a reply content; Encrypting the generated reply content according to the second secret key to obtain the reply information; Sending the reply information to the client, The first key and the second key are pre-set keys used for encrypted transmission of the current session.

15. The method according to claim 9, wherein The generating, by the target customer service, and sending a reply message to the client based on the received session information includes: In response to the session information including information related to a historical session, obtaining at least a portion of a session record of the historical session; Based on at least a portion of the session record of the historical session and the session information, a reply message is generated by the target customer service and sent to the client.

16. A session management device for customer service scenarios, comprising: a session initiation module configured to: in response to receiving a session initiation operation of an object, send a session initiation request to a server, wherein the session initiation request is used to initiate a current conversation and includes an object identifier of the object using the client; a receiving module configured to: receive feedback information regarding the session initiation request from the server, wherein the feedback information is configured to indicate that a target customer service has been determined, wherein the target customer service is a customer service that matches the object and is determined based on the object identifier; a first sending module configured to: in response to receiving a conversation input operation of the object, send conversation information to the server, wherein the conversation information includes conversation content input by the object through the conversation input operation; a presentation module configured to: in response to receiving reply information from the server, present the reply information, wherein the reply information is generated by the target customer service based on the session information; The second sending module is configured to: in response to the object exiting the current session, send information indicating the end of the current session to the server, so that when the server receives the information indicating the end of the current session and determines that a session deletion condition is met, delete at least a part of the session record of the current session, and the session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction from the object for at least a part of the session record of the current session is received.

17. A session management device for customer service scenarios, comprising: A first receiving module is configured to: receive a session initiation request from a client, wherein the session initiation request is used to initiate a current conversation and includes an object identifier of an object using the client; A determination module is configured to: determine a target customer service that matches the object based on the object identifier; a sending module configured to: send feedback information regarding the session initiation request to the client, wherein the feedback information is configured to indicate that the target customer service has been determined; A second receiving module is configured to: receive session information from the client, wherein the session information includes session content input by the subject through the client; A reply module is configured to: generate and send reply information to the client through the target customer service based on the received session information; The deletion module is configured to: in response to the current session ending and satisfying the session deletion condition, delete at least a part of the session record of the current session, wherein the session deletion condition includes at least one of the following: the current session involves privacy data, the current session belongs to a privacy session scenario, and a deletion instruction is received from the object for at least a part of the session record of the current session.

18. A computing device comprising: a memory configured to store computer-executable instructions; A processor configured to perform the method according to any one of claims 1 to 15 when the computer executable instructions are executed by the processor.

19. A computer-readable storage medium storing computer-executable instructions, which, when executed, perform the method according to any one of claims 1 to 15.

20. A computer program product comprising computer executable instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 15.

Images