User interaction method and device, equipment, medium and program product

Abstract

The invention provides a user interaction method and device, equipment, a medium and a program product, which can be applied to the technical field of artificial intelligence. The method comprises the following steps: when a user accesses a system, constructing a cache data set based on behaviors of the user in the system; acquiring behavior data of the user from the cache data set; obtaining historical behavior characteristics of the user from a vector database of the system through semantic matching based on the behavior data; obtaining structured record data of the user recorded in the system from a structured database of the system; inputting the behavior data, the historical behavior characteristics and the structured record data into a large language model, and predicting potential services intended to be operated by the user based on a plurality of services which can be subsequently provided by the system by utilizing the large language model; and recommending an interaction entrance of the potential service to the user.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence and can be used in the field of fintech or other fields, and more specifically to a user interaction method, apparatus, device, medium and program product. Background Technology

[0002] Current online business user interaction systems are typically based on pre-developed user interfaces, which are highly versatile but lack personalization. They generally lack effective long-term memory and deep personalized service capabilities. Summary of the Invention

[0003] The present invention provides a user interaction method, apparatus, device, medium, and program product that can at least partially solve the above-mentioned problems.

[0004] A first aspect of this invention provides a user interaction method. The method includes: when a user accesses a system, constructing a cached dataset based on the user's behavior in the system; obtaining the user's behavior data from the cached dataset; obtaining the user's historical behavior features from the system's vector database through semantic matching based on the behavior data; retrieving the user's structured record data from the system's structured database based on the behavior data; inputting the behavior data, the historical behavior features, and the structured record data into a large language model, using the large language model to predict potential services for the user's intended operation based on multiple services subsequently available from the system; and recommending interaction entry points for the potential services to the user.

[0005] According to an embodiment of the present invention, constructing a cached dataset based on a user's behavior in the system includes: recording the user's behavior in the system in real time into the cached dataset; wherein the behavior data includes behavior sequences.

[0006] According to an embodiment of the present invention, constructing a cache dataset based on a user's behavior in the system further includes: analyzing the user's behavioral preferences based on the user's behavior in the system; and recording the user's behavioral preferences into the cache dataset; wherein the behavioral data includes the behavioral preferences.

[0007] According to an embodiment of the present invention, the method further includes: updating the user's behavioral preferences in the cached dataset based on the user's processing result of the interaction entry, wherein the processing result includes whether the user operated the interaction entry or did not operate the interaction entry.

[0008] According to an embodiment of the present invention, the method further includes: periodically or by an event triggering, extracting the user's behavioral features from the cached dataset, wherein the event includes the user exiting the system or the amount of data in the cached dataset exceeding a threshold; and storing the behavioral features in the vector database.

[0009] According to an embodiment of the present invention, the plurality of services are those that require multiple steps to access from the current interaction location.

[0010] According to an embodiment of the present invention, the method further includes: when receiving an operation from the user on the interaction entry point, directly jumping to the potential service.

[0011] A second aspect of the present invention provides a user interaction device. The device includes a caching module, a prediction module, and an interaction module.

[0012] The caching module is used to build a cached dataset based on the user's behavior in the system when the user accesses the system.

[0013] The prediction module is used to obtain the user's behavioral data from the cached dataset; obtain the user's historical behavioral features from the system's vector database through semantic matching based on the behavioral data; obtain the user's structured record data recorded in the system from the system's structured database; and input the behavioral data, the historical behavioral features, and the structured record data into a large language model, and use the large language model to predict the potential services for the user's intended operation based on multiple services that the system can subsequently provide.

[0014] The interaction module is used to recommend the potential services to the user.

[0015] According to an embodiment of the present invention, the caching module is specifically used to: record the user's behavior in the system in real time to the cache dataset.

[0016] According to an embodiment of the present invention, the caching module is specifically used for: analyzing the user's behavioral preferences based on the user's behavior in the system; and recording the user's behavioral preferences into the cache dataset.

[0017] According to an embodiment of the present invention, the caching module is further configured to: update the user's behavioral preferences in the cached dataset based on the user's processing result of the interaction entry, wherein the processing result includes whether the user operated the interaction entry or did not operate the interaction entry.

[0018] According to an embodiment of the present invention, the caching module is further configured to: extract the user's behavioral features from the cached dataset at regular intervals or triggered by an event, wherein the event includes the user exiting the system or the amount of data in the cached dataset exceeding a threshold; store the behavioral features in the vector database; and clear the cached dataset in response to the completion of the behavioral feature storage.

[0019] According to an embodiment of the present invention, the interaction module is further configured to: when receiving an operation from the user on the interaction entry point, directly jump to the potential service.

[0020] A third aspect of the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0021] A fourth aspect of the present invention also provides a computer-readable storage medium having a computer program or instructions stored thereon, wherein the computer program or instructions, when executed by a processor, implement the steps of the above-described method.

[0022] A fifth aspect of the present invention also provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps of the above-described method.

[0023] The above one or more embodiments have the following advantages or beneficial effects: behavioral data obtained from the cached dataset reflects the user's current behavioral characteristics; historical behavioral characteristics matched semantically from the vector database reflect long-term behavioral characteristics related to the current behavior; and structured record data of the user retrieved from the structured database can provide relevant background information for analyzing the user's current behavior. Thus, when the large language model predicts the user's current potential needs based on these three inputs, it can combine long-term memory and the user's current needs to make a comprehensive judgment, thereby providing personalized and targeted services and improving the accuracy of the prediction. Attached Figure Description

[0024] The above-described features, other objects, and advantages of the present invention will become clearer from the following description of embodiments of the invention with reference to the accompanying drawings, in which:

[0025] Figure 1 The illustrations depict application scenarios of user interaction methods, apparatuses, devices, media, and program products according to embodiments of the present invention.

[0026] Figure 2 A flowchart illustrating a user interaction method according to an embodiment of the present invention is shown schematically;

[0027] Figure 3 The system architecture of a user interaction method according to an embodiment of the present invention is illustrated schematically;

[0028] Figure 4 This illustration schematically shows the process of constructing a cached dataset in a user interaction method according to an embodiment of the present invention;

[0029] Figure 5 A flowchart illustrating a user interaction method according to another embodiment of the present invention is shown schematically; and

[0030] Figure 6 A schematic diagram illustrating the structure of a user interaction device according to an embodiment of the present invention is shown; and

[0031] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing a user interaction method according to an embodiment of the present invention. Detailed Implementation

[0032] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the invention. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the invention for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0033] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0034] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0035] In the technical solution of this invention, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0036] In scenarios where personal information is used for automated decision-making, the methods, devices, and systems provided in this embodiment of the invention all provide users with corresponding operation entry points for users to choose to agree to or reject the automated decision results; if the user chooses to reject, the process proceeds to the expert decision-making process.

[0037] Figure 1 The illustration shows an application scenario diagram of the user interaction method, apparatus, device, medium, and program product according to embodiments of the present invention.

[0038] like Figure 1 As shown, application scenario 100 according to this embodiment may include at least one terminal device (three are shown in the figure: a first terminal device 101, a second terminal device 102, and a third terminal device 103), a network 104, and a server 105. Network 104 is a medium used to provide a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0039] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, financial applications, e-banking, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0040] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0041] Server 105 may be a server that provides various services, such as a back-end management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (for example only).

[0042] Server 105 can execute the user interaction method of this embodiment of the invention. Based on the user's access behavior in the system supported by server 105 (such as a bank back-end service system, which is only an example), it predicts the potential service that the user intends to operate on and proactively recommends the interaction entry point of the potential service to the user, thereby improving the intelligence of user interaction. Accordingly, the user interaction device, electronic device, medium, and program product provided in this embodiment of the invention can be set in server 105.

[0043] It should be noted that the user interaction method provided in the embodiments of the present invention can also be executed by a server or server cluster that is different from server 105 and is capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or server 105. Accordingly, the user interaction device, electronic device, medium and program product provided in the embodiments of the present invention can also be located in a server or server cluster that is different from server 105 and is capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or server 105.

[0044] It should be understood that Figure 1 The examples shown are merely illustrative of application scenarios for embodiments of the present invention, and the application scenarios of the embodiments of the present invention are not limited thereto. Furthermore, the user interaction methods, devices, equipment, media, and program products determined in the embodiments of the present invention can be used in the fintech field, as well as in any field other than fintech; the present invention does not limit the application field.

[0045] The following will be based on Figure 1 The described application scenarios, through Figures 2-5 The user interaction method according to embodiments of the present invention will be described in detail.

[0046] Figure 2 A flowchart illustrating a user interaction method according to an embodiment of the present invention is shown.

[0047] like Figure 2 As shown, the user interaction method of this embodiment may include operations S210 to S260, and this user interaction method may be executed by server 105. (In conjunction with...) Figure 3 The system architecture is described below.

[0048] First, in operation S210, when a user accesses the system, a cached dataset 301 is constructed based on the user's behavior in the system. The user's behavior in the system includes, but is not limited to: various operations on the user interface of the system (searching, viewing, filling out forms, shopping or placing orders, etc.), or communication and dialogue with customer service in the system.

[0049] In one embodiment, the construction process of cache dataset 301 can be as follows: Figure 4 As shown, this includes operations S401 to S403.

[0050] In operation S401, the user's behavior in the system is recorded in real time to the cache dataset 301. For example, the cache dataset 301 records how long the user stays on each page or screen, what information they view, which controls they operate, the operation area information on the page or screen (such as a virtual interactive screen), or the content of their conversation with online customers.

[0051] In addition, in operation S402, user behavior preferences can be analyzed based on the user's behavior in the system. Then, in operation S403, the user's behavior preferences are recorded in the cache dataset 301. User behavior preferences may include, but are not limited to: which area the user focuses on in viewing information, whether the user quickly browses or carefully reads various pages or screens, which content the user carefully reads and thus infers the user's interest in, and whether the user's operation area is concentrated in the left-hand or right-hand area and thus the user's current gesture preferences can be analyzed.

[0052] Next, in operation S220, the user's behavioral data is obtained from the cached dataset 301. This behavioral data may include behavioral sequences obtained from the user's real-time operation data, or behavioral preferences derived from analysis.

[0053] Then, in operation S230, the user's historical behavioral characteristics are obtained from the system's vector database 302 through semantic matching based on the behavioral data.

[0054] And in operation S240, the user's structured record data is retrieved from the system's structured database 303 based on the behavioral data.

[0055] Next, in operation S250, the behavioral data, the historical behavioral features, and the structured record data are input into the large language model 304. The large language model 304 is then used to predict the potential services for the user's intended operation based on multiple services that the system can subsequently provide.

[0056] In some embodiments, the large language model 304 can be obtained by fine-tuning a pre-trained large language model, such as by fine-tuning and learning using the content of a large number of services provided by the system, the flow relationships or branch relationships between services. In this way, the large language model 304 can guide users to the corresponding functional branches in the system by predicting potential services. In other embodiments, the large language model 304 can also use the content of a large number of services provided by the external server system, the flow relationships or branch relationship data between services, and retrieve relevant data to predict services when predicting user intent.

[0057] Then, in operation S260, the interactive entry point of the potential service is recommended to the user. If the user interacts with the interactive entry point, the user is redirected to the interactive node of the potential service, and the potential service is provided to the user.

[0058] The behavioral data obtained from the cached dataset 301 reflects the user's current behavioral characteristics, while the historical behavioral features matched semantically from the vector database 302 reflect long-term behavioral characteristics related to the current behavior. The structured record data of the user retrieved from the structured database 303 can provide relevant background information for analyzing the user's current behavior. Thus, the large language model 304 can analyze the input, predict the user's current potential needs, and combine long-term memory with the user's current needs to provide personalized and targeted services, improving the accuracy of predictions.

[0059] It should be noted that before obtaining user behavior data in operation S220, obtaining user historical behavior characteristics in operation S230, and obtaining user structured record data in operation S240, the user's consent or authorization can be obtained. For example, before operation S220, an authorization request for using the relevant data can be sent to the user. If the user agrees or authorizes the acquisition of the relevant data for interaction optimization, then operations S220, S230, and S240 can be executed.

[0060] In one embodiment, the multiple services predicted by the large language model 304 in operation S250 are services that require multiple steps to access from the current interaction position. Accordingly, if the user's operation on the interaction entry point is received, the user can be directly redirected to the potential service, allowing the user to quickly obtain the required service.

[0061] According to an embodiment of the present invention, after operation S260, the user's behavioral preferences in the cached dataset 301 are updated based on the user's processing result of the interaction entry point, wherein the processing result includes whether the user operated the interaction entry point or not. Specifically, if the user operated the interaction entry point, it indicates that the judgment of the user's operation intention was correct, thereby improving the credibility of the analyzed behavioral preferences. For example, the behavioral preferences can be stored in a vector database for permanent preservation. If the user did not operate the interaction entry point, it indicates that the judgment of the user's operation intention was incorrect. In this case, the credibility of the previously analyzed behavioral preferences can be reduced, or the previously analyzed behavioral preferences can be changed according to the user's new operation action, so that the large language model 304 can make more accurate predictions based on the new behavioral preferences.

[0062] In this embodiment of the invention, the large language model 304 analyzes and predicts the user's operation intention. This can be done when the user visits certain key pages in the system or communicates with customer service, or it can be based on user behavior, such as making recommendations when the user repeatedly jumps between multiple pages or multiple controls, or it can be done at regular intervals.

[0063] Next, refer to Figure 3 The user interaction method of this embodiment of the invention is further described below. In this embodiment of the invention, the system deployed in server 105 adopts a dual-database operating architecture, including a vector database 302 and a structured database 303. Server 105 can communicate with both the vector database 302 and the structured database 303, which can provide data support for the operation of the system in server 105.

[0064] Structured databases (303 statements) can store structured data. Taking a bank's back-end service system as an example, a structured database (303 statement) can store user profiles, such as basic information, account information, asset data, transaction records, or risk tags.

[0065] Vector database 302 can store various unstructured data, such as historical dialogue text, user preference expressions, customer service response styles, and sentiment analysis results. Specifically, the collected raw unstructured data can be vectorized through dialogue semantic embedding and behavioral pattern vectorization to form storable and semantically recognizable data, which can then be stored in vector database 302.

[0066] During the user's access to the system, behavioral data obtained from the cached dataset 301, historical behavioral features matched from the vector database 302, and the user's structured record data retrieved from the structured database 303 can be input into the large language model 304, which will then predict the user's next operational intent.

[0067] The behavioral data obtained from the cached dataset 301 reflects the user's current behavioral characteristics. The historical behavioral features matched semantically from the vector database 302 reflect long-term behavioral patterns related to the current behavior. The structured record data of the user retrieved from the structured database 303 can provide relevant background information for analyzing the user's current behavior. The large language model 304 can analyze the input and predict the user's current potential needs.

[0068] For example, the input to the Big Language Model 304 is structured data (such as age / assets / risk assessment) and unstructured data (current and historical conversation records / operational behaviors). Big Language Model 304 can then combine knowledge graph technology to associate user attributes with behavioral tags (such as "conservative - prefers bond funds - frequently checks returns"), and further combine this with behavioral data extracted from cached datasets (assuming the user is browsing various funds) to predict the user's current implicit needs (such as "wanting to save for their child's education"), thereby inferring that the user wants to view services related to saving for education. This allows the interaction entry points of services corresponding to the user's implicit needs to be recommended to the user, for example, by displaying an area corresponding to the user's current gesture preferences.

[0069] As users access the system, the cached dataset 301 will continuously expand. Correspondingly, valuable information from the cached dataset 301 will be persisted to the vector database 302, and the cached dataset will be cleared in a timely manner. One relevant embodiment can be found here. Figure 5 .

[0070] like Figure 5 As shown, the user interaction method according to this embodiment may include operations S510 to S530 in addition to operations S210 to S260.

[0071] In operation S510, the user's behavioral features are extracted from the cached dataset 301 at regular intervals or triggered by events. These events include the user exiting the system or the amount of data in the cached dataset 301 exceeding a threshold. Feature extraction can be performed using clustering algorithms or algorithms such as term frequency-inverse document frequency (TF-IF).

[0072] In operation S520, the behavioral features are stored in the vector database 302.

[0073] In operation S530, in response to the completion of storage of the behavioral features, the cached dataset 301 is cleared.

[0074] After storing behavioral features in vector database 302, if the user continues to access the system, the existing data in cached dataset 301 can be cleared and re-accumulated. If the user no longer accesses the system, cached dataset 301 is released.

[0075] The embodiments of the present invention can establish a unified long-term memory for structured and unstructured data. By combining a structured database 303 that stores structured data (such as user transaction records, holdings, asset allocation, and market conditions) and a vector database 302 that stores unstructured data (such as historical dialogue text, user preference expressions, customer service response styles, and sentiment analysis results), a cross-time dimension "memory" foundation is provided for user interaction in the system.

[0076] In this embodiment of the invention, the system can achieve deep personalized interaction based on long-term memory. Relying on the multidimensional database's ability to perform multidimensional correlation analysis on user current behavior data, historical behavior characteristics, and user profiles (such as user portrait data) stored in the structured database 303, the system can predict potential user needs by combining historical context and current behavior.

[0077] The system in this embodiment of the invention can achieve progressive understanding of the current interaction context. It can associate the current interaction behavior with historical behavior data in the vector database 302 to understand the user's long-term concerns, unresolved issues, and emotional trajectory, thus making the prediction of user intent more coherent and in-depth.

[0078] The system described in this invention can intelligently mine user preferences. It can dynamically adjust recommendation strategies based on analysis of user preferences and changes in current interactions, combined with preferences exhibited in historical interactions, thus balancing the long-term nature of recommendations with timely adaptation to the current situation.

[0079] The method of this invention can be applied to customer service interaction scenarios. When a user uses the system's customer service dialogue, the large language model 304 can predict the user's potential needs based on long-term behavioral data, interaction history, and current behavioral data recorded in the cached dataset 301 stored and associated in a multi-dimensional database, and make targeted recommendations. This allows for the dynamic construction and updating of more refined and comprehensive user profiles, driving intelligent customer service to simulate the personalized service style and interaction capabilities of a human customer manager. Specifically, when using the system's customer service dialogue, the system can predict the user's potential needs based on behavioral data, historical behavioral characteristics, and structured record data, and form targeted scripts and interactive entry points for recommending potential services. In some embodiments, the generated style can also be dynamically adjusted based on the user's communication language (such as the rate of liking humorous statements), supporting preset personality templates: professional and rigorous (citing report data) or friendly and humorous (using metaphors to defuse fluctuations).

[0080] Based on the user interaction methods of the above embodiments, this invention also provides a user interaction device. The following will be combined with... Figure 6 The device is described in detail.

[0081] Figure 6 A schematic block diagram of a user interaction device 600 according to an embodiment of the present invention is shown.

[0082] like Figure 6 As shown, the user interaction device 600 according to this embodiment includes: a caching module 601, a prediction module 602, and an interaction module 603.

[0083] The caching module 601 is used to construct a cached dataset 301 based on the user's behavior in the system when the user accesses the system. In one embodiment, the caching module 601 may perform the operation S210 described above.

[0084] The prediction module 602 is used to obtain the user's behavioral data from the cached dataset 301; based on the behavioral data, obtain the user's historical behavioral features from the system's vector database 302 through semantic matching; obtain the user's structured record data recorded in the system from the system's structured database 303; and input the behavioral data, the historical behavioral features, and the structured record data into a large language model 304, using the large language model 304 to predict the potential service for the user's intended operation based on multiple services that the system can subsequently provide. In one embodiment, the prediction module 602 can perform the operations S220 to S250 described above.

[0085] The interaction module 603 is used to recommend the interactive entry point of the potential service to the user. In one embodiment, the interaction module 603 is further used to: directly jump to the potential service when it receives an operation from the user on the interactive entry point. In one embodiment, the interaction module 603 can perform the operation S260 described above.

[0086] In one embodiment, the caching module 601 is further configured to: extract the user's behavioral features from the cached dataset 301 at regular intervals or triggered by an event, wherein the event includes the user exiting the system or the amount of data in the cached dataset 301 exceeding a threshold; and store the behavioral features in the vector database 302.

[0087] The user interaction device 600 can perform reference Figures 2-5 The user interaction methods introduced are described in detail above and will not be repeated here.

[0088] According to embodiments of the present invention, any plurality of modules among the cache module 601, prediction module 602, and interaction module 603 may be combined into one module, or any one of these modules may be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules may be combined with at least a portion of the functionality of other modules and implemented in one module. According to embodiments of the present invention, at least one of the cache module 601, prediction module 602, and interaction module 603 may be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the cache module 601, prediction module 602, and interaction module 603 may be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0089] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing a user interaction method according to an embodiment of the present invention.

[0090] like Figure 7 As shown, an electronic device 900 according to an embodiment of the present invention includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0091] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 executes various operations of the method flow according to embodiments of the present invention by executing programs in ROM 902 and / or RAM 903. It should be noted that the programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also execute various operations of the method flow according to embodiments of the present invention by executing programs stored in said one or more memories.

[0092] According to an embodiment of the present invention, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0093] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0094] According to embodiments of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of the present invention, a computer-readable storage medium may include ROM 902 and / or RAM 903 and / or one or more memories other than ROM 902 and RAM 903 described above.

[0095] Embodiments of the present invention also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the methods provided in the embodiments of the present invention.

[0096] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this invention. According to embodiments of the invention, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0097] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 909, and / or installed from a removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0098] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, it performs the functions defined in the system of this embodiment of the invention. According to embodiments of the invention, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0099] According to embodiments of the present invention, program code for executing the computer programs provided in the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0101] Those skilled in the art will understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention can be combined and / or combined in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or combinations fall within the scope of the present invention.

Claims

1. A user interaction method, comprising: constructing a cache dataset based on a user’s behavior in a system when the user accesses the system; obtaining behavior data of the user from the cache dataset; obtaining historical behavior features of the user from a vector database of the system through semantic matching based on the behavior data; retrieving structured record data of the user from a structured database of the system based on the behavior data; inputting the behavior data, the historical behavior features and the structured record data into a large language model, and predicting potential services that the user intends to operate based on a plurality of services that the system can subsequently provide using the large language model; and recommending an interaction portal of the potential services to the user. Constructing a cache dataset based on a user’s behavior in a system comprises:

2. The method of claim 1, wherein, recording the user’s behavior in the system to the cache dataset in real time; wherein the behavior data comprises a behavior sequence. Constructing a cache dataset based on a user’s behavior in a system further comprises:

3. The method of claim 2, wherein, analyzing the user’s behavior preferences based on the user’s behavior in the system; and recording the user’s behavior preferences to the cache dataset; wherein the behavior data comprises the behavior preferences. The method further comprises:

4. The method of claim 3, wherein, updating the user’s behavior preferences in the cache dataset based on the user’s processing result of the interaction portal, wherein the processing result comprises operating the interaction portal or not operating the interaction portal. The method further comprises:

5. The method of claim 1, wherein, extracting behavior features of the user from the cache dataset at a timing or triggered by an event, wherein the event comprises the user logging out of the system or the amount of data in the cache dataset being greater than a threshold; and storing the behavior features to the vector database; and emptying the cache dataset. The plurality of services are services that require multiple steps of jumping from a current interaction location to enter.

6. The method of claim 1, wherein, The method further comprises:

7. The method of claim 1 or 6, wherein, directly jumping to the potential services when receiving the user’s operation on the interaction portal.

8. A user interaction apparatus, comprising: a cache module configured to construct a cache dataset based on a user’s behavior in a system when the user accesses the system; a prediction module configured to: obtain behavior data of the user from the cache dataset; obtain historical behavior features of the user from a vector database of the system through semantic matching based on the behavior data; obtain structured record data of the user recorded in the system from a structured database of the system; and input the behavior data, the historical behavior features and the structured record data into a large language model, and predict potential services that the user intends to operate based on a plurality of services that the system can subsequently provide using the large language model; an interaction module configured to recommend an interaction portal of the potential services to the user.

9. An electronic device, comprising: one or more processors; a memory configured to store one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-7. ​ ​ ​ 10. A computer readable storage medium having stored thereon a computer program or instructions, wherein, The computer program or instructions, when executed on a processor, implement the steps of the method according to any one of claims 1-7.

11. A computer program product comprising computer programs or instructions, wherein, The computer program or instructions, when executed on a processor, implement the steps of the method according to any one of claims 1-7.

Images