Interaction method and device and electronic equipment

Abstract

The invention discloses an interaction method and device and electronic equipment, and the method comprises the steps: determining initial privacy information corresponding to a target user according to the user information of the target user, the initial privacy information being used for limiting the range of callable information in a target information base; acquiring first information corresponding to a target user acquired by an information acquisition device; adjusting the initial privacy information into target privacy information according to the first information; in response to the received question information of the target user, determining reference information from the target information base according to the target privacy information; and determining response information corresponding to the question information according to the reference information.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to an interaction method, device and electronic device. Background Art

[0002] With the rapid development of Internet technology, personalized services have become an important means to improve user experience. When providing personalized services, the system needs to access and process users' personal information to meet their specific needs. However, how to effectively protect user privacy while ensuring service quality has become an urgent problem to be solved. Traditional privacy management methods usually use fixed privacy settings to manage access rights to user information. Although this method is simple and easy to implement, it often fails to strike a good balance between personalization and privacy protection in practical applications. On the one hand, overly strict privacy control may limit the system's ability to obtain necessary information and affect service quality; on the other hand, if the privacy setting is too loose, it may lead to the leakage of sensitive information and increase security risks. Summary of the invention

[0003] The present application embodiment provides an interactive method, comprising:

[0004] Determine the initial privacy information corresponding to the target user according to the user information of the target user, wherein the initial privacy information is used to limit the scope of information that can be called in the target information database;

[0005] Acquire first information corresponding to the target user collected by the information collection device;

[0006] adjusting the initial private information to target private information according to the first information;

[0007] In response to the received question information of the target user, determining reference information from the target information database according to the target private information;

[0008] Response information corresponding to the question information is determined according to the reference information.

[0009] Optionally, adjusting the initial private information to target private information according to the first information includes:

[0010] Determining, according to the first information, that there is a first user in the target space where the target user is located, and the first user is different from the target user;

[0011] Determining association information between the target user and the first user;

[0012] The initial private information is adjusted to the target private information according to the association information.

[0013] Optionally, the association information includes one of the following:

[0014] Identity relationship information between the target user and the first user;

[0015] Communication behavior information between the target user and the first user.

[0016] Optionally, adjusting the initial private information to the target private information according to the association information includes:

[0017] Determining communication content information from the communication behavior information;

[0018] According to the matching relationship between the communication content information and the information contained in the target information library, the initial private information is adjusted to the target private information.

[0019] Optionally, the target information database includes an information sub-database corresponding to the target user personal information; and adjusting the initial private information to the target private information according to the association information includes:

[0020] The initial private sub-information corresponding to the information sub-database is adjusted to the target private sub-information according to the identity relationship information; the initial private information includes the initial private sub-information, and the target private information includes the target private sub-information.

[0021] Optionally, the initial privacy information includes an initial privacy level corresponding to the target user, the target privacy information includes a target privacy level corresponding to the target user, and adjusting the initial privacy information to the target privacy information according to the first information includes:

[0022] The initial privacy level is adjusted to the target privacy level according to the first information.

[0023] Optionally, the initial private information includes callable status information corresponding to each of the plurality of pieces of information in the target information library, and the callable status information is used to determine whether the corresponding information can be called based on the question information; and adjusting the initial private information to the target private information according to the first information includes:

[0024] The callable state information corresponding to at least one of the multiple pieces of information is adjusted according to the first information to obtain the target private information.

[0025] Optionally, the first information corresponding to the target user collected by the information collection device includes at least one of the following:

[0026] Image information corresponding to the target user collected by the image collection device;

[0027] Voice information corresponding to the target user collected by the audio collection device;

[0028] The radar signal information corresponding to the target user collected by the radar signal collection device.

[0029] Optionally, determining the response information corresponding to the question information according to the reference information includes:

[0030] Determine target prompt information according to the question information and the reference information;

[0031] The target prompt information is sent to the target model to obtain the response information generated by the target model.

[0032] The embodiment of the present application also provides an interactive device, including:

[0033] A first determination module is configured to determine initial privacy information corresponding to the target user according to the user information of the target user, wherein the initial privacy information is used to limit the scope of information that can be called in the target information database;

[0034] An acquisition module configured to acquire first information corresponding to a target user acquired by an information acquisition device;

[0035] an adjustment module, configured to adjust the initial private information to target private information according to the first information;

[0036] A second determination module is configured to determine reference information from the target information database according to the target privacy information in response to the received question information of the target user;

[0037] The third determining module is configured to determine the answer information corresponding to the question information according to the reference information.

[0038] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein the memory stores an executable program, and the processor executes the executable program to perform the steps of the method described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flowchart of an interactive method according to an embodiment of the present application;

[0040] Figure 2 Embodiments of the present application Figure 1 A flowchart of an embodiment of step S300;

[0041] Figure 3 This is a structural block diagram of an interactive device according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] Various aspects and features of the present application are described herein with reference to the accompanying drawings.

[0043] It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the present application will occur to those skilled in the art.

[0044] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0045] These and other characteristics of the present application will become apparent from the following description of a preferred form of embodiment given as a non-limiting example with reference to the accompanying drawings.

[0046] It should also be understood that although the present application has been described with reference to some specific examples, those skilled in the art will be able to readily implement many other equivalent forms of the present application.

[0047] The above and other aspects, features and advantages of the present application will become more apparent in view of the following detailed description when taken in conjunction with the accompanying drawings.

[0048] Specific embodiments of the present application are described hereinafter with reference to the accompanying drawings; however, it should be understood that the embodiments applied for are merely examples of the present application, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that obscure the present application. Therefore, the specific structural and functional details applied for herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present application in a variety of ways with substantially any suitable detailed structure.

[0049] This specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present application.

[0050] In related technologies, although some systems try to improve the situation of fixed privacy settings through RBAC (Role-Based Access Control) or user-defined privacy settings, these methods generally have the problem of insufficient flexibility. For example, once the privacy permissions are set, it is difficult to dynamically adjust the access level according to changes in user behavior. In addition, in question-answering systems, when facing users with different backgrounds and needs, how to properly filter out information that not only meets their personal preferences but also does not violate privacy principles is also a major challenge.

[0051] Therefore, it is necessary to develop a new interactive method that can intelligently adjust privacy settings according to the specific circumstances of the user and provide more accurate and secure services on this basis. The interactive method provided in this application automatically adjusts the privacy level through real-time analysis of user information, and can achieve efficient information retrieval and service response under the premise of protecting user privacy.

[0052] The interactive method of the present application is described in detail below with reference to the accompanying drawings. Figure 1 is a flow chart of the interaction method of an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0053] S100: Determine initial privacy information corresponding to a target user according to user information of the target user, where the initial privacy information is used to limit the scope of information that can be called in a target information database.

[0054] In an embodiment of the present application, the user information of the target user includes, but is not limited to, the registration information, historical interaction records, device information used, geographic location information, identity information, action information, voice information, expression information, emotional information, tone information, health status information, etc. of the target user in communication with other users. The collected user information can be used to construct a user profile of the target user. Among them, the user profile may include key features such as user preferences, behavior patterns, and privacy sensitivity. Optionally, the initial privacy information may represent that all information in the target information library can be called. Optionally, the initial privacy information may represent that all information corresponding to the target user in the target information library can be called. Optionally, the initial privacy information may represent the permission level corresponding to the user identity of the target user.

[0055] The information in the target information database may include information that the user agrees to share, information that the user does not agree to share, and information that can be shared under specific conditions. Among them, the information that the user agrees to share may be name, gender, and age, etc. The information that the user does not agree to share may be financial information and health information, etc. Information that can be shared under specific conditions refers to information that can only be accessed, used, or disseminated if certain preset conditions or restrictions are met. For example, in a social media platform, users can choose in the settings whether their posts are publicly visible or only visible to friends or specific groups, and can also choose whether to allow third-party applications to access their personal information and content through an API (Application Programming Interface). For another example, in a learning management system, a student's transcript and evaluation report can be configured to be viewed only by the student himself and the teacher.

[0056] In an embodiment of the present application, initial privacy information can be used to limit the scope of information access and protect sensitive data. As the first line of defense for the interaction between the user and the intelligent system, the initial privacy information determines which information in the target information library can be queried or used. For example, if a user has a high degree of privacy sensitivity, the system can only allow access to the user's basic public information and shield their detailed personal data. In addition, through preset rules such as identity authentication and access control, data classification and label setting, it can be ensured that the leakage of sensitive information can be effectively prevented even before any personalized adjustments are made.

[0057] S200: Acquire first information corresponding to the target user collected by the information collection device.

[0058] In the embodiment of the present application, obtaining the first information corresponding to the target user collected by the information collection device includes at least one of the following:

[0059] The image information corresponding to the target user collected by the image collection device; wherein the image collection device can be a camera, a fingerprint collector, etc.

[0060] The voice information corresponding to the target user collected by the audio collection device; wherein the audio collection device may be a microphone.

[0061] The radar signal information corresponding to the target user collected by the radar signal collection device; wherein the radar signal collection device can be a handheld or portable radar, or a vehicle-mounted radar, an airborne radar, etc.

[0062] It is understandable that the information collection device can be a hardware device or a software component. Among them, the user's biometric information and physical environment information can be directly obtained through the hardware device. For example, the user's fingerprint features can be collected through a fingerprint identifier, the user's facial features can be collected through a facial recognition camera, and the user's voiceprint features can be collected through a microphone. The user's geographic location information can also be collected through a location sensor.

[0063] Information about users’ activities in digital environments can be collected through software components, which can be logging tools within applications, browser cookies, and web tracking codes. Information about users’ activities in digital environments can include clickstream data, page dwell time, browsing habits, interaction patterns, and frequency of use.

[0064] Of course, context information can also be obtained through hardware devices or software components, such as the current time and device status, etc. It is also possible to obtain relevant information in specific application scenarios, such as sensitive information such as financial information and health information.

[0065] It can be understood that the first information corresponding to the target user is the original data obtained directly from the target user through the information collection device, which can be the user's biometric information, physical environment information, activity information in the digital environment, contextual information or other sensitive information corresponding to the specific application scenario.

[0066] In addition, the collection of the first information corresponding to the target user can be automatically started when a specific event occurs, or can be actively triggered by the user. For example, when a user logs into an application, the application collects their location information; or when a user makes a voice query, the microphone is activated to capture the user's audio data.

[0067] S300, adjusting the initial private information to target private information according to the first information;

[0068] In an embodiment of the present application, first, the obtained first information is preprocessed, specifically including: cleaning and standardizing the first information, removing noise and invalid data, and ensuring data quality. Secondly, key features are extracted from the preprocessed first information. These key features may be the user's geographic location, behavior pattern, or frequency of use. For example, the geographic location can be used to identify whether the user is in a public space or a private space. Thirdly, the preprocessed first information is contextually analyzed in combination with the user's historical data and other relevant data to obtain a contextual analysis result. Finally, based on the key features and the contextual analysis results, the user's current status and needs are obtained. The initial privacy information is then adjusted according to the user's current status and needs.

[0069] In an embodiment of the present application, the scope of accessible information in the target database can be adjusted in real time based on the collected user information, thereby preventing the risk of data leakage or insufficient data calls due to changes in the target user's environment causing the scope of currently accessible information to be incompatible, thereby achieving dynamic adjustment of the access scope of the information department in the target database.

[0070] In an embodiment of the present application, the adjustment of the initial privacy information to the target privacy information can be achieved through a preset rule template or a machine learning model. Specifically, based on the preset rule template, the privacy settings of the initial privacy information are automatically adjusted according to the current status and needs of the user. For example, if the user is detected to be in a public place, the user's privacy protection level is increased. The machine learning model can also be trained with historical data so that the model continuously learns the user's behavior patterns to predict the best privacy settings. Of course, multiple factors such as time, location, device type, and user behavior can also be considered to determine the privacy settings; through a weighted scoring mechanism, a weight is assigned to each factor to calculate the final privacy level and obtain the final target privacy information.

[0071] It is understandable that by continuously collecting the first information corresponding to the user in real time, combined with an efficient event processing mechanism and a low-latency data transmission method, dynamic adjustment of the initial privacy information can be achieved. In addition, a user feedback interface can be provided to allow the user to manually adjust the privacy settings or confirm the system's automatic adjustment of the privacy settings, thereby helping to improve user experience and trust.

[0072] S400: In response to the received question information of the target user, determine reference information from the target information database according to the target private information.

[0073] In an embodiment of the present application, a user's question information can be received through a user interface, wherein the user interface can be a chat window or a voice assistant, etc., and the user's question information can be in text form or in voice form. The target privacy information can be used for privacy level checking and data screening. Specifically, before processing the user's question, the current target privacy information is first checked to determine the user's information access rights; then, based on the target privacy information, the data that is allowed to be accessed is screened from the target information library. For example, if the user's privacy setting is high, the system only accesses public or non-sensitive information.

[0074] In an embodiment of the present application, information retrieval techniques such as full-text search, vector space model, inverted index, etc. can be used to search for information related to the user's question from the filtered data set; and contextual information such as the user's historical interaction records and geographic location can be combined to improve the relevance of the search results. For example, if the user has previously asked about a medical topic, the system gives priority to data related to this. Then, the information found that is related to the user's question can be scored for relevance, and the information with the highest score can be used as reference information.

[0075] S500: Determine response information corresponding to the question information according to the reference information.

[0076] In the embodiment of the present application, first, the reference information needs to be processed, including extracting key content from the reference information, and further semantically analyzing the extracted key content using natural language processing technology to determine its relevance to the user's question. Among them, the key content extracted from the reference information may include facts, data, descriptions, and definitions. Then, a preset response template is used to fill in the response template in combination with the user's question and reference information to obtain the response information corresponding to the question information.

[0077] It is understandable that in order to ensure the consistency and coherence of the response information, it is also necessary to combine the user's historical questions and historical interaction records to construct the response information. For example, if the user has mentioned background information about medical care in the previous question, the response information can continue to be constructed on this basis. In addition, the content and form of the response information can also be adjusted according to the user's preferences and behavior patterns. For example, for users who like concise answers, a shorter response information can be provided; for users who need detailed explanations, response information with more background information can be provided.

[0078] The interactive method provided in the embodiment of the present application can adjust the privacy settings of the user according to the real-time information of the user, and provide personalized question-and-answer services on this basis. In this way, not only can the user experience be effectively improved, but also the security of user information can be ensured.

[0079] In one embodiment, in the above step S300, the initial private information is adjusted to the target private information according to the first information, such as Figure 2 As shown, including:

[0080] S310: Determine, according to the first information, that a first user exists in the target space where the target user is located, and the first user is different from the target user.

[0081] S320: Determine association information between the target user and the first user.

[0082] S330: Adjust the initial private information to the target private information according to the association information.

[0083] In this embodiment, the space where the target user is located can be determined by using a camera, microphone, Bluetooth or Wi-Fi signal strength, and then face recognition and voice recognition methods can be used to determine whether there are other users in the same space.

[0084] A relationship map containing the relationships between users can be pre-built, where the relationships between users include but are not limited to social relationships, work relationships, etc. The relationship information can be quantified by intimacy and trust level, etc. The privacy protection level can be adjusted based on the relationship, combined with preset security rules or personal preferences.

[0085] In this embodiment, when the first user is the same as the target user, that is, when there is only one target user in the target space, the privacy settings can be adjusted according to the user's actions, voice, tone, etc. Specifically, the user's body movements can be identified through camera acquisition and image recognition technology. For example, if the user is identified to make a face-covering action, it means that the user may want to increase privacy protection. The user's privacy needs can also be identified by monitoring specific words or phrases in the user's voice content. For example, if it is identified that "quiet is needed now" in the user's voice content, the device volume is lowered or the notification is stopped. The user's emotional state can also be judged through emotional analysis of the voice, thereby inferring his privacy needs. For example, if the system detects that the user's voice is nervous or uneasy, consider enhancing the privacy settings.

[0086] In this embodiment, the association information includes identity relationship information between the target user and the first user, or communication behavior information between the target user and the first user.

[0087] When the association information is identity relationship information between the target user and the first user, the target information database is preset with an information sub-database corresponding to the target user's personal information; and adjusting the initial private information to the target private information according to the association information includes:

[0088] The initial private sub-information corresponding to the information sub-database is adjusted to the target private sub-information according to the identity relationship information; the initial private information includes the initial private sub-information, and the target private information includes the target private sub-information.

[0089] Specifically, the target information database can be divided into different sub-databases according to personal information, and each sub-database corresponds to a specific type of information, such as work information, family information, health information, and financial information. Access rights can be set for each information sub-database, and different identity relationships correspond to different access levels. Each information sub-database has corresponding initial privacy settings, which can preset which information can be accessed and which cannot be accessed by default.

[0090] By setting up multiple information sub-libraries in the target information library, users' personal information can be managed at a finer granularity. Each information sub-library contains a specific type of information and has an initial privacy setting. According to the identity relationship between the user and other people, such as family, friends, colleagues, etc., the system automatically adjusts the access rights of these information sub-libraries and adjusts the initial privacy sub-information to the target privacy sub-information. This dynamic adjustment not only takes into account the relationship between users, but also combines the current situational factors, making privacy management more intelligent and personalized. At the same time, a user feedback mechanism can be set up to allow users to view and confirm these adjustments to ensure that the privacy settings meet the actual needs of users. In this way, it is possible to protect user privacy while ensuring effective information sharing.

[0091] In addition, when the associated information is the communication behavior information between the target user and the first user, adjusting the initial private information to the target private information according to the associated information includes:

[0092] Determining communication content information from the communication behavior information;

[0093] According to the matching relationship between the communication content information and the information contained in the target information library, the initial private information is adjusted to the target private information.

[0094] Specifically, first, natural language processing technology can be used to extract communication content information from communication behavior information. Secondly, the communication content information is classified, for example, work-related, personal life-related, and health-related, and corresponding labels can be configured for each classification to facilitate subsequent processing and matching. Thirdly, for each type of communication content information, the keywords and topics in the communication content are matched with the corresponding information in the target information library, and the relevant matching content with a matching relationship is determined in the target information library, and the accuracy of the matching is improved through the context information of the communication behavior information. Finally, based on the matching relationship, the initial privacy information is adjusted to the target privacy information to meet the current communication needs. In this way, the user's privacy settings can be managed intelligently, which can not only protect the security of personal information, but also promote effective information sharing.

[0095] In this embodiment, when the associated information is the communication behavior information between the target user and the first user, adjusting the initial private information to the target private information according to the associated information may further include:

[0096] Determining action information during communication from the communication behavior information;

[0097] According to the matching relationship between the action information during the communication and the action patterns preset in the target information library and the meanings or situations that they may imply, the initial private information is adjusted to the target private information.

[0098] Specifically, for example, if it is detected that the user frequently blocks the screen or hides input during a conversation, it means that the user is discussing a sensitive topic and the protection level of privacy information needs to be increased.

[0099] In addition, when the associated information is the communication behavior information between the target user and the first user, adjusting the initial private information to the target private information according to the associated information may further include:

[0100] Determining tone information during communication from the communication behavior information;

[0101] According to the matching relationship between the tone information during the communication and the preset tone type in the target information library and its corresponding context or privacy setting, the initial private information is adjusted to the target private information.

[0102] Specifically, the tone information during communication includes but is not limited to different types of tones such as positive, negative, formal, informal, urgent, and relaxed. For example, if the tone during communication is formal and cautious, it means that the user may be discussing a more sensitive topic; on the contrary, if it is casual and friendly, it means that the user may be discussing something less important. Based on the matching results, the user's privacy information is adjusted accordingly. Specifically, if the analysis of the user's tone shows that the current conversation involves sensitive topics, the protection level of the privacy information is increased, for example, the visibility of certain content is restricted.

[0103] It should be noted that, in actual applications, based on specific needs, the initial privacy information can be adjusted to the target privacy information based on one of the communication content information, action information during communication and tone information during communication determined from the communication behavior information; or the initial privacy information can be adjusted to the target privacy information based on a fusion result of the communication content information, action information during communication and tone information during communication determined from the communication behavior information.

[0104] In another embodiment, in the above step S300, the initial privacy information includes an initial privacy level corresponding to the target user, the target privacy information includes a target privacy level corresponding to the target user, and adjusting the initial privacy information to the target privacy information according to the first information includes:

[0105] The initial privacy level is adjusted to the target privacy level according to the first information.

[0106] Specifically, a privacy level system can be established, and a default initial privacy level can be preset for each piece of user information; when multiple information sub-libraries are set in the target information library, a default initial privacy level can also be preset for each information sub-library. For example, the privacy level system includes levels 1-5, where level 1 represents the highest privacy protection and level 5 represents the lowest privacy protection. In this way, the initial privacy information is adjusted to the target privacy information by adjusting the privacy level.

[0107] In another embodiment, in the above step 300, the initial private information includes callable status information corresponding to each of the plurality of pieces of information in the target information library, and the callable status information is used to determine whether the corresponding information can be called based on the question information; and adjusting the initial private information to the target private information according to the first information includes:

[0108] The callable state information corresponding to at least one of the multiple pieces of information is adjusted according to the first information to obtain the target private information.

[0109] Specifically, the information in the target information database can be divided into multiple items, and initial callable status information can be set for each information item to determine whether the information can be called based on the question information. In addition, preset adjustment rules can adjust the callable status information of each information item in the target information database according to the results of environmental perception, identity relationship recognition and communication content analysis.

[0110] In another embodiment, in the above step S500, determining the response information corresponding to the question information according to the reference information includes:

[0111] Determine target prompt information according to the question information and the reference information;

[0112] The target prompt information is sent to the target model to obtain the response information generated by the target model.

[0113] Specifically, natural language processing technology can be used to parse the target user's question information to extract keywords, topics and intentions. First, the information that is allowed to be accessed is filtered out from the target information library based on the current target privacy information of the target user, and the content related to the question extracted from the information that is allowed to be accessed is used as reference information. Then, a target prompt information containing a complete context is generated by combining the question information and the reference information. Then, the target prompt information is sent to a preset target model, which processes the input target prompt information and generates corresponding response information. Finally, the generated response information can also be quality checked and provided to the user. In addition, a user feedback mechanism can be set up to collect user feedback to continuously optimize the target model and the generation process of the response information, thereby improving the quality and accuracy of the response information.

[0114] Optionally, the target model is a large language model or a multimodal large model. Optionally, the target model can generate an output result corresponding to any natural language instruction received. Optionally, the number of adjustable parameters of the target model is greater than 100 million.

[0115] Based on the same inventive concept, the present application embodiment also provides an interactive device, such as Figure 3 As shown, including:

[0116] The first determination module is configured to determine initial privacy information corresponding to the target user according to the user information of the target user, wherein the initial privacy information is used to limit the scope of information that can be called in the target information library.

[0117] The acquisition module is configured to acquire the first information corresponding to the target user collected by the information collection device.

[0118] An adjustment module is configured to adjust the initial private information to target private information according to the first information.

[0119] The second determination module is configured to determine reference information from the target information database according to the target privacy information in response to the received question information of the target user.

[0120] The third determining module is configured to determine the answer information corresponding to the question information according to the reference information.

[0121] In the embodiment of the present application, the user information of the target user includes but is not limited to the registration information, historical interaction records, device information used, and geographic location information of the target user. The collected user information can be used to construct a user profile of the target user. The user profile may include key features such as the user's preferences, behavior patterns, and privacy sensitivity.

[0122] Initial privacy information includes information that the user agrees to share, information that the user does not agree to share, and information that can be shared under specific conditions. Among them, the information that the user agrees to share can be name, gender, and age, etc. The information that the user does not agree to share can be financial information and health information, etc. Information that can be shared under specific conditions refers to information that can only be accessed, used, or disseminated if certain preset conditions or restrictions are met. For example, in social media platforms, users can choose in the settings whether their posts are publicly visible or only visible to friends or specific groups, and can also choose whether to allow third-party applications to access their personal information and content through the API (Application Programming Interface). For another example, in a learning management system, students' transcripts and evaluation reports can be configured to be viewed only by the students themselves and their teachers.

[0123] In an embodiment of the present application, initial privacy information can be used to limit the scope of information access and protect sensitive data. As the first line of defense for the interaction between users and intelligent systems, initial privacy information determines which information in the target information library can be queried or used. For example, if a user has a high degree of privacy sensitivity, the system can only allow access to the user's basic public information and shield their detailed personal data. In addition, through preset rules, such as identity authentication and access control, data classification and label settings, etc., it can be ensured that the leakage of sensitive information can be effectively prevented even before any personalized adjustments are made.

[0124] In the embodiment of the present application, obtaining the first information corresponding to the target user collected by the information collection device includes at least one of the following:

[0125] The image information corresponding to the target user collected by the image collection device; wherein the image collection device can be a camera, a fingerprint collector, etc.

[0126] The voice information corresponding to the target user collected by the audio collection device; wherein the audio collection device may be a microphone.

[0127] The radar signal information corresponding to the target user collected by the radar signal collection device; wherein the radar signal collection device can be a handheld or portable radar, or a vehicle-mounted radar, an airborne radar, etc.

[0128] It is understandable that the information collection device can be a hardware device or a software component. Among them, the user's biometric information and physical environment information can be directly obtained through the hardware device. For example, the user's fingerprint features can be collected through a fingerprint identifier, the user's facial features can be collected through a facial recognition camera, and the user's voiceprint features can be collected through a microphone. The user's geographic location information can also be collected through a location sensor.

[0129] Information about users’ activities in digital environments can be collected through software components, which can be logging tools within applications, browser cookies, and web tracking codes. Information about users’ activities in digital environments can include clickstream data, page dwell time, browsing habits, interaction patterns, and frequency of use.

[0130] Of course, context information can also be obtained through hardware devices or software components, such as the current time and device status, etc. It is also possible to obtain relevant information in specific application scenarios, such as sensitive information such as financial information and health information.

[0131] It can be understood that the first information corresponding to the target user is the original data obtained directly from the target user through the information collection device, which can be the user's biometric information, physical environment information, activity information in the digital environment, contextual information or other sensitive information corresponding to the specific application scenario.

[0132] In addition, the collection of the first information corresponding to the target user can be automatically started when a specific event occurs, or can be actively triggered by the user. For example, when a user logs into an application, the application collects their location information; or when a user makes a voice query, the microphone is activated to capture the user's audio data.

[0133] In an embodiment of the present application, first, the obtained first information is preprocessed, specifically including: cleaning and standardizing the first information, removing noise and invalid data, and ensuring data quality. Secondly, key features are extracted from the preprocessed first information. These key features may be the user's geographic location, behavior pattern, or frequency of use. For example, the geographic location can be used to identify whether the user is in a public space or a private space. Thirdly, the preprocessed first information is contextually analyzed in combination with the user's historical data and other relevant data to obtain a contextual analysis result. Finally, based on the key features and the contextual analysis results, the user's current status and needs are obtained. The initial privacy information is then adjusted according to the user's current status and needs.

[0134] In an embodiment of the present application, the adjustment of the initial privacy information to the target privacy information can be achieved through a preset rule template or a machine learning model. Specifically, based on the preset rule template, the privacy settings of the initial privacy information are automatically adjusted according to the current status and needs of the user. For example, if the user is detected to be in a public place, the user's privacy protection level is increased. The machine learning model can also be trained with historical data so that the model continuously learns the user's behavior patterns to predict the best privacy settings. Of course, multiple factors such as time, location, device type, and user behavior can also be considered to determine the privacy settings; through a weighted scoring mechanism, a weight is assigned to each factor to calculate the final privacy level and obtain the final target privacy information.

[0135] It is understandable that by continuously collecting the first information corresponding to the user in real time, combined with an efficient event processing mechanism and a low-latency data transmission method, dynamic adjustment of the initial privacy information can be achieved. In addition, a user feedback interface can be provided to allow users to manually adjust privacy settings or confirm the system's automatic adjustment, thereby helping to improve user experience and trust.

[0136] In an embodiment of the present application, a user's question information can be received through a user interface, wherein the user interface can be a chat window or a voice assistant, etc., and the user's question information can be in text form or in voice form. The target privacy information can be used for privacy level checking and data screening. Specifically, before processing the user's question, the intelligent system first checks the current target privacy information to determine the user's information access rights; then, based on the target privacy information, the data that is allowed to be accessed is screened from the target information library. For example, if the user's privacy setting is high, the system only accesses public or non-sensitive information.

[0137] In an embodiment of the present application, information retrieval techniques such as full-text search, vector space model, inverted index, etc. can be used to search for information related to the user's question from the filtered data set; and contextual information such as the user's historical interaction records and geographic location can be combined to improve the relevance of the search results. For example, if the user has previously asked about a medical topic, the system gives priority to data related to this. Then, the information found that is related to the user's question can be scored for relevance, and the information with the highest score can be used as reference information.

[0138] In the embodiment of the present application, first, the reference information needs to be processed, including extracting key content from the reference information, and further semantically analyzing the extracted key content using natural language processing technology to determine its relevance to the user's question. Among them, the key content extracted from the reference information may include facts, data, descriptions, and definitions. Then, a preset response template is used to fill in the response template in combination with the user's question and reference information to obtain the response information corresponding to the question information.

[0139] It is understandable that in order to ensure the consistency and coherence of the response information, the response information needs to be constructed based on the user's historical questions and historical interaction records. For example, if the user has mentioned background information about medical care in the previous question, the response information can continue to be constructed on this basis. In addition, the content and form of the response information can also be adjusted according to the user's preferences and behavior patterns. For example, for users who like concise answers, a shorter response information can be provided; for users who need detailed explanations, response information with more background information can be provided.

[0140] The interactive method provided in the embodiment of the present application can adjust the privacy settings of the user according to the real-time information of the user, and provide personalized question-and-answer services on this basis. In this way, not only can the user experience be effectively improved, but also the security of user information can be ensured.

[0141] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein the memory stores an executable program, and the processor executes the executable program to perform the steps of the method described above.

[0142] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. An interactive method, comprising: Determine the initial privacy information corresponding to the target user according to the user information of the target user, wherein the initial privacy information is used to limit the scope of information that can be called in the target information database; Acquire first information corresponding to the target user collected by the information collection device; adjusting the initial private information to target private information according to the first information; In response to the received question information of the target user, determining reference information from the target information database according to the target private information; Response information corresponding to the question information is determined according to the reference information.

2. The method according to claim 1, wherein adjusting the initial private information to target private information according to the first information comprises: Determining, according to the first information, that there is a first user in the target space where the target user is located, and the first user is different from the target user; Determining association information between the target user and the first user; The initial private information is adjusted to the target private information according to the association information.

3. The method according to claim 2, wherein the association information comprises one of the following: Identity relationship information between the target user and the first user; Communication behavior information between the target user and the first user.

4. The method according to claim 3, wherein adjusting the initial private information to the target private information according to the association information comprises: Determining communication content information from the communication behavior information; According to the matching relationship between the communication content information and the information contained in the target information library, the initial private information is adjusted to the target private information.

5. The method according to claim 3, wherein the target information database has an information sub-database corresponding to the target user personal information; and adjusting the initial private information to the target private information according to the association information comprises: Adjusting the initial private sub-information corresponding to the information sub-database to target private sub-information according to the identity relationship information; The initial private information includes the initial private sub-information, and the target private information includes the target private sub-information.

6. The method according to claim 1, wherein the initial privacy information includes an initial privacy level corresponding to the target user, the target privacy information includes a target privacy level corresponding to the target user, and the adjusting the initial privacy information to the target privacy information according to the first information comprises: The initial privacy level is adjusted to the target privacy level according to the first information.

7. The method according to claim 1, wherein the initial private information includes callable status information corresponding to each of the plurality of pieces of information in the target information library, and the callable status information is used to determine whether the corresponding information can be called based on the question information; and adjusting the initial private information to the target private information according to the first information comprises: The callable state information corresponding to at least one of the multiple pieces of information is adjusted according to the first information to obtain the target private information.

8. According to the method of claim 1, the first information corresponding to the target user collected by the information collection device includes at least one of the following: Image information corresponding to the target user collected by the image collection device; Voice information corresponding to the target user collected by the audio collection device; The radar signal information corresponding to the target user collected by the radar signal collection device.

9. The method according to claim 1, wherein determining the response information corresponding to the question information according to the reference information comprises: Determine target prompt information according to the question information and the reference information; The target prompt information is sent to the target model to obtain the response information generated by the target model.

10. An interactive device, comprising: A first determination module is configured to determine initial privacy information corresponding to the target user according to the user information of the target user, wherein the initial privacy information is used to limit the scope of information that can be called in the target information database; An acquisition module configured to acquire first information corresponding to a target user acquired by an information acquisition device; an adjustment module, configured to adjust the initial private information to target private information according to the first information; A second determination module is configured to determine reference information from the target information database according to the target privacy information in response to the received question information of the target user; The third determining module is configured to determine the answer information corresponding to the question information according to the reference information.

11. An electronic device, comprising a memory and a processor, wherein an executable program is stored in the memory, and the processor executes the executable program to perform the steps of the method according to any one of claims 1 to 9.

Images