Information processing method and device, computer readable medium and electronic equipment

Abstract

The invention discloses an information processing method and device, a computer readable medium and electronic equipment, and the method comprises the steps: carrying out the retrieval in a preset database according to user input information, and obtaining database retrieval information matched with the user input information; performing information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information; generating task prompt information according to the user input information, the database retrieval information and the perception enhancement information; and generating target result information corresponding to the user input information according to the task prompt information. According to the technical scheme, the associated information of the user input information is expanded through the preset database retrieval and the information perception enhancement processing, so that when the target result information is finally generated, the database retrieval dimension and the information perception enhancement dimension can be further considered; therefore, the intelligent agent can more comprehensively grasp the related multi-dimensional information input by the user, and the accuracy of data processing is improved.

Description

Technical Field

[0001] The present application belongs to the field of artificial intelligence technology, and specifically relates to an information processing method, device, computer-readable medium and electronic device. Background Art

[0002] Agent systems refer to systems composed of multiple autonomous or semi-autonomous entities (called agents) that can perceive, reason and take actions in a specific environment to achieve a predetermined goal. In traditional agent systems, agents only process their own input information and then output the processing result data, which results in the data dimensions considered when the agent generates the processing result are not comprehensive enough, and the accuracy of data processing needs to be improved.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] The purpose of this application is to provide an information processing method, device, computer-readable medium and electronic device to improve the accuracy of data processing in an intelligent agent system.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to one aspect of an embodiment of the present application, there is provided an information processing method, including:

[0007] Searching in a preset database according to the user input information to obtain database search information matching the user input information;

[0008] Performing information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information;

[0009] generating task prompt information according to the user input information, the database search information and the perception enhancement information;

[0010] Generate target result information corresponding to the user input information according to the task prompt information.

[0011] According to one aspect of an embodiment of the present application, there is provided an information processing device, including:

[0012] A search module, used to search in a preset database according to user input information to obtain database search information matching the user input information;

[0013] an enhancement module, configured to perform information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information;

[0014] a prompt generation module, configured to generate task prompt information according to the user input information, the database search information and the perception enhancement information;

[0015] A processing module is used to generate target result information corresponding to the user input information according to the task prompt information.

[0016] In one embodiment of the present application, the retrieval module includes:

[0017] A first retrieval unit, configured to perform vectorization processing on the user input information to obtain a user input vector, and search the preset database according to the user input vector to obtain first retrieval information;

[0018] A second retrieval unit, configured to extract keywords from the user input information, and search the preset database according to the keywords to obtain second retrieval information;

[0019] An integration unit is used to integrate the first search information and the second search information to generate database search information matching the user input information.

[0020] In one embodiment of the present application, the data types in the preset database include knowledge data, historical interaction data, historical agent selection data and thought arrangement data; the first retrieval unit is specifically used to:

[0021] Determine data of each data type that matches the user input vector according to the similarity between the user input vector and the data vectors corresponding to each data type in the preset database;

[0022] The first search information is obtained according to data of each data type that matches the user input vector.

[0023] In one embodiment of the present application, the second retrieval unit is specifically used to:

[0024] Obtain an inverted index corresponding to each data type in the preset database, wherein the inverted index indicates an index relationship between a data keyword and the data in which it is located;

[0025] Search the inverted index corresponding to each data type according to the keyword to determine the data corresponding to each data type that contains the keyword;

[0026] The second search information is obtained according to the data containing the keyword corresponding to each data type.

[0027] In one embodiment of the present application, the integration unit is specifically used for:

[0028] Deduplication processing is performed on data of each data type that matches the user input vector and data corresponding to each data type that contains the keyword to obtain database search information that matches the user input information.

[0029] In one embodiment of the present application, the enhancement module is specifically used to:

[0030] Performing notification enhancement processing on the user input information in terms of time dimension, space dimension and vertical domain information dimension respectively, to obtain time enhancement information, space enhancement information and vertical domain enhancement information respectively;

[0031] At least one of the temporal enhancement information, the spatial enhancement information and the vertical field enhancement information is used as the perception enhancement information corresponding to the user input information.

[0032] In one embodiment of the present application, the processing module is specifically used to:

[0033] Calling a designated agent to process the user input information according to the task prompt information to obtain agent output information;

[0034] When the output condition is not met, the agent output information is added to the preset database, and the process returns to the step of searching the preset database according to the user input information;

[0035] When the output condition is met, the agent output information is used as the target result information corresponding to the user input information; wherein, satisfying the output condition includes that the number of times the agent processes the user input information reaches a preset number of iterations, or the evaluation index corresponding to the agent output information meets the index requirement.

[0036] According to one aspect of an embodiment of the present application, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the information processing method provided by any embodiment of the present application is implemented.

[0037] According to one aspect of an embodiment of the present application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor executes the executable instructions so that the electronic device executes the information processing method provided by any embodiment of the present application.

[0038] According to one aspect of an embodiment of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the information processing method provided by any embodiment of the present application.

[0039] In the technical solution provided in the embodiment of the present application, firstly, a search is performed in a preset database according to the user input information to obtain database search information matching the user input information; at the same time, information perception enhancement processing is performed on the user input information to obtain perception enhancement information corresponding to the user input information; then task prompt information is generated according to the user input information, database search information and perception enhancement information; finally, target result information corresponding to the user input information is generated according to the task prompt information. In this way, the associated information of the user input information is expanded through the preset database search and information perception enhancement processing, so that when the target result information is finally generated, the database search dimension and the information perception enhancement dimension can be further considered, so that the intelligent agent can more comprehensively grasp the multi-dimensional information related to the user input, and improve the accuracy of data processing.

[0040] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 The exemplary system architecture block diagram applying the technical solution of the present application is schematically shown.

[0043] Figure 2 The flowchart of an information processing method provided by an embodiment of the present application is schematically shown.

[0044] Figure 3 The flowchart of an information processing method provided by an embodiment of the present application is schematically shown.

[0045] Figure 4 The flowchart of an information processing method provided by an embodiment of the present application is schematically shown.

[0046] Figure 5The structural block diagram of the information processing device provided in an embodiment of the present application is schematically shown.

[0047] Figure 6 The structure block diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0048] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0049] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0050] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0051] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0052] Figure 1 The exemplary system architecture block diagram applying the technical solution of the present application is schematically shown.

[0053] like Figure 1As shown, the system architecture 100 may include a terminal device 110, a network 120 and a server 130. The terminal device 110 may include a smart phone, a tablet computer, a laptop computer, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, etc. The server 130 may be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The network 120 may be a communication medium of various connection types that can provide a communication link between the terminal device 110 and the server 130, for example, a wired communication link or a wireless communication link.

[0054] The technical solution of the present application can promote the development of intelligent services and serve the specific needs of different industries, especially large network models, including but not limited to: operational decision-making, fault reconstructing, on-site comprehensive maintenance, installation and maintenance knowledge, rules and regulations and other scenarios. According to the implementation needs, the system architecture in the embodiment of the present application can have any number of terminal devices, networks and servers. For example, the server 130 can be a server group composed of multiple server devices. In addition, the technical solution provided in the embodiment of the present application can be applied to the terminal device 110, can also be applied to the server 130, or can be implemented by the terminal device 110 and the server 130 together, and the present application does not make any special restrictions on this.

[0055] The information processing method provided by the present application is described in detail below in conjunction with specific implementation methods.

[0056] Figure 2 The flowchart of an information processing method provided by an embodiment of the present application is schematically shown. The method can be implemented by an intelligent agent system, and the intelligent agent system can be run on a terminal device or a server. Figure 2 As shown, the method comprises the following steps:

[0057] Step 210: Search in a preset database according to the user input information to obtain database search information matching the user input information.

[0058] Specifically, the preset database includes historical interaction behavior data of users in the agent system. These historical interaction behavior data may include multiple types of data, such as knowledge data, historical interaction data, historical agent selection data, thought arrangement data, etc. Knowledge data refers to the knowledge information used in the process of historical interaction with users, and may also include knowledge information that may be involved in processing user input information. In some cases, the agent system may be a system for processing problems in a certain technical field, and the knowledge data may include knowledge information involved in the technical field. Historical interaction data is data generated when users interact in the agent system, such as a question-and-answer dialogue between users and the agent system. Historical interaction data is usually text data, and may also include data of types such as images, codes, and videos. The agent system may be a system composed of multiple agents. In the process of interacting with users, only some agents may be used. The agent data selected in the user interaction process is the historical agent selection data. In some cases, in the process of interacting with users, agents usually use multiple tools to process user tasks (user tasks are generated based on user input information), such as translation tools, search tools, etc. The selection data of these tools can also be used as historical agent selection data. Thinking orchestration data refers to a set of processes used by the intelligent system when processing user tasks. Thinking orchestration data is usually used to enable data in the user task processing process to flow efficiently and reliably between different storage, processing and services in the intelligent system, and to be correctly accessed and used when needed.

[0059] By searching the preset database according to the user input information, database search information matching the user input information can be obtained, and the database search information can be any one or more data of the above multiple data types. By constructing the preset database of the above multiple data types, a comprehensive context information can be formed for the user input information, and the obtained database search information fully considers the context environment of the user, thereby enhancing the personalized service and context understanding ability of subsequent data processing.

[0060] Step 220: Perform information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information.

[0061] Specifically, information perception enhancement processing refers to the multi-dimensional expansion of user input information to obtain data with richer dimensions. This can improve the intelligent system's ability to understand data, thereby more accurately and comprehensively capturing and interpreting user input information.

[0062] In one embodiment of the present application, information perception enhancement processing can be performed on the user input information from at least one of the time dimension, the space dimension and the vertical domain information dimension, so that the perception enhancement information corresponding to the user input information includes at least one of the time enhancement information, the space enhancement information and the vertical domain enhancement information.

[0063] Information perception enhancement for the time dimension includes but is not limited to date query, holiday query, weather query, time difference query, etc., so that the agent system can process user input information according to time information. Exemplarily, for the time dimension, the current timestamp of the user input information can be obtained as its time enhancement information.

[0064] For information perception enhancement in the spatial dimension, including but not limited to geocoding, map services, spatial analysis tools, navigation and path planning tools, GIS (Geographic Information System) integration tools, etc., the increase in spatial dimensions enables the intelligent system to better understand and respond to user needs related to geographic location and provide richer and more personalized services. For example, for the spatial dimension, the current geographic location of the user input information can be obtained as its spatial enhancement information.

[0065] For the information perception enhancement of the vertical domain information dimension, it is applied to the vertical domain intelligent agent scenario. Taking the network big model as an example, it can include network status perception, network topology perception, network security perception, network performance perception, etc., so that the network big model intelligent agent can better understand and adapt to the network environment and provide more efficient, secure and personalized services. Exemplarily, for the vertical domain information dimension, the stability of the user's current network can be obtained as its vertical domain enhancement information.

[0066] Step 230: Generate task prompt information based on user input information, database retrieval information and perception enhancement information.

[0067] Specifically, task prompt information refers to a piece of text or instruction provided to the agent system to guide the agent system to generate a response to the user input information. By adding database retrieval information and perception enhancement information to the task prompt information, the data dimension in the task prompt information is effectively expanded, thereby improving the quality and relevance of the agent output, and thus obtaining the target result information corresponding to the user input information more quickly.

[0068] In one embodiment of the present application, when constructing task prompt information, in addition to user input information, database retrieval information and perception enhancement information, data such as the agent's configuration file, user-specified behavior guidelines, tool specifications, and predefined output formats may also be added. The agent's configuration file contains the agent's configuration information. The user-specified behavior guidelines include the user's requirements for the agent system to process input information, such as reflecting the calculation process. Tool specifications include specifications of various large language models (LLMs) used by the agent system, such as the model's network size, structural hierarchy, etc. The predefined output format represents the data format of the target result information ultimately output by the agent system.

[0069] Step 240: Generate target result information corresponding to the user input information according to the task prompt information.

[0070] Specifically, the task prompt information and the user input information can be simultaneously input into the intelligent agent system, and the intelligent agent system calls the corresponding large language model for processing, and outputs the target result information corresponding to the user input information.

[0071] In the technical solution provided in the embodiment of the present application, firstly, a search is performed in a preset database according to the user input information to obtain database search information matching the user input information; at the same time, information perception enhancement processing is performed on the user input information to obtain perception enhancement information corresponding to the user input information; then task prompt information is generated according to the user input information, database search information and perception enhancement information; finally, target result information corresponding to the user input information is generated according to the task prompt information. In this way, the associated information of the user input information is expanded through the preset database search and information perception enhancement processing, so that when the target result information is finally generated, the database search dimension and the information perception enhancement dimension can be further considered, so that the intelligent agent can more comprehensively grasp the multi-dimensional information related to the user input, and improve the accuracy of data processing.

[0072] Figure 3 The flowchart of the information processing method provided by one embodiment of the present application is schematically shown. This embodiment is a further refinement of the above embodiment. Figure 3 As shown, the information processing method provided in the embodiment of the present application includes the following steps:

[0073] Step 310: vectorize the user input information to obtain a user input vector, and search in a preset database according to the user input vector to obtain first search information.

[0074] Specifically, an embedded vector search method can be used to search in a preset database. In this search method, the user input information is first vectorized to obtain a user input vector, and each data in the preset database also needs to be vectorized to obtain each data vector. Then the data vector that matches the user input vector is retrieved, and the data corresponding to the data vector is used as the first search information. For example, the similarity or distance between the user input vector and each data vector can be calculated, and the data vector with high similarity or distance can be used as the target data vector that matches the user input vector. The data vector with high similarity or distance can be a data vector corresponding to the maximum similarity or distance, or it can be one or more data vectors with a similarity or distance higher than a threshold.

[0075] In one embodiment of the present application, based on the knowledge data, historical interaction data, historical agent selection data, and thought arrangement data contained in the preset database, an embedded vector search can be performed for each type of data, so that data matching the user input vector can be obtained for each data type; finally, the data matching the user input vector for all data types can be used as the first search information. In the process of vectorizing the data in the preset database, the data of each data type can be first divided into text segments of fixed length, and then the text segments can be vectorized.

[0076] Step 320: extract keywords from the user input information, and search in a preset database according to the keywords to obtain second search information.

[0077] Specifically, you can use the inverted index search method to search in the preset database. Before searching, you must first extract the keywords of each data in the database, and then build an inverted index of each data keyword and the data in the preset database. The forward index is to find the words in the document through the document, and the inverted index is the opposite. It determines the document where it is located through the vocabulary. In the search process, first extract keywords from the user input information, and then match the extracted keywords in the inverted index to find the consistent data keywords. The data corresponding to the data keyword is the second search information.

[0078] In one embodiment of the present application, based on the knowledge data, historical interaction data, historical agent selection data and thinking arrangement data contained in the preset database, an inverted index search can be performed for each type of data separately, so that the data containing the user input information keywords corresponding to each data type can be obtained; finally, the data containing the user input information keywords corresponding to all data types can be used as the second search information.

[0079] In one embodiment of the present application, when searching for data keywords matching user input information keywords, the search can be performed by string matching, or both the user input information keywords and the data keywords can be vectorized and then searched by vector matching.

[0080] Step 330: Integrate the first search information and the second search information to generate database search information matching the user input information.

[0081] Specifically, the first search information and the second search information obtained based on the above two search methods may have overlapping data, so they are deduplicated, and only one of the overlapping data is retained. Finally, the deduplicated data are integrated together as database search information matching the user input information.

[0082] In one embodiment of the present application, during the deduplication process, if one of the multiple repeated data is the only data of a certain data type in the search information, then the repeated item can retain the data of the data type, so that it can be ensured that in the final database search information, each type of data has its corresponding data, so that the distribution of each type of data is relatively uniform. Exemplarily, it is assumed that the first search information and the second search information include knowledge data 1, knowledge data 2, historical interaction data 1, historical intelligent agent selection data 1, historical intelligent agent selection data 2, thinking arrangement data 1 and thinking arrangement data 2, among which knowledge data 1, historical interaction data 1, historical intelligent agent selection data 1 and thinking arrangement data 1 are repeated data. Considering that historical interaction data 1 is the only data of the data type of historical interaction data, therefore, historical interaction data 1 is retained and the remaining repeated data is deleted, and the database search information finally obtained includes: knowledge data 2, historical interaction data 1, historical intelligent agent selection data 2 and thinking arrangement data 2.

[0083] In one embodiment of the present application, the database retrieval information M integrated It can be expressed as: M integrated =f(M knowledge , M conversation , M tool , M thinking ), where M knowledge represents the knowledge data retrieval results, M conversation represents the historical interaction data retrieval results, M tool represents the historical agent selection data retrieval result, M thinking Indicates thinking about arranging data retrieval results.

[0084] Step 340: Perform notification enhancement processing on the user input information in terms of time dimension, space dimension and vertical domain information dimension, respectively, to obtain time enhancement information, space enhancement information and vertical domain enhancement information, respectively.

[0085] Step 350: Use at least one of the temporal enhancement information, the spatial enhancement information, and the vertical domain enhancement information as the perception enhancement information corresponding to the user input information.

[0086] Step 360: Generate task prompt information based on user input information, database retrieval information and perception enhancement information.

[0087] The specific implementation process of step 340-step 360 can refer to the relevant description in the aforementioned step 220-step 230, which will not be repeated here.

[0088] In one embodiment of the present application, the task prompt information T prompt It can be expressed as: prompt =g(Q,M integrated , P), where Q represents user input information, M integrated represents database retrieval information, and P represents agent configuration information, user-specified behavior guidelines, and perception enhancement information.

[0089] Step 370: Call the designated agent to process the user input information according to the task prompt information to obtain the agent output information.

[0090] Specifically, when processing user input information, the intelligent agent system can generate a user task based on the task prompt information and the user input information, and then generate at least one user task command, and then call the corresponding intelligent agent or large language model to execute the user task to obtain the intelligent agent output information.

[0091] In one embodiment of the present application, the agent output information O can be expressed as: O = h(T command , M integrated ), where M integrated Represents database retrieval information, T command Represents a user task command, which is a subset of user tasks.

[0092] In one embodiment of the present application, the agent output information can be directly used as the target result information corresponding to the user input information.

[0093] Step 380: When the output condition is not met, the agent output information is added to the preset database and the process returns to step 310.

[0094] Step 390: When the output conditions are met, the agent output information is used as the target result information corresponding to the user input information; wherein, meeting the output conditions includes that the number of times the agent processes the user input information reaches a preset number of iterations, or the evaluation index corresponding to the agent output information meets the index requirements.

[0095] Specifically, the agent output information is continuously updated in an iterative manner so that the target result information finally obtained is more accurate. In this process, if the output condition is not met, it is necessary to continuously iterate. During the iteration process, the agent output information is added to the preset database to update the preset database in real time so that the next output can refer to the previous output result to improve the accuracy of the final result. When the output condition is met, for example, the number of times the agent processes the user input information reaches the preset number of iterations, or the evaluation index corresponding to the agent output information meets the index requirements, then the agent output information obtained at the last time is the target result information corresponding to the user input information.

[0096] In one embodiment of the present application, the target result information R can be expressed as: R = k(Q, M integrated , P, O), where Q represents user input information, M integrated represents database retrieval information, P represents agent configuration information, user-specified behavior guidelines and perception enhancement information, and O represents the last obtained agent output information.

[0097] In the technical solution provided in the embodiments of the present application, data matching the user input information is retrieved from multiple aspects through vector-based embedded vector retrieval and keyword-based inverted index retrieval, thereby improving the accuracy of the retrieval; through information enhancement processing in at least one dimension such as time, space and vertical field information, the data processing dimension is effectively expanded, and the thinking and decision-making ability of the intelligent agent is improved; finally, the target result information is obtained in an iterative manner, which improves the autonomy of the intelligent agent and reduces the dependence on human intervention, so that the intelligent agent can respond to changes and perform tasks more quickly and independently.

[0098] Figure 4 A flowchart of an information processing method provided by an embodiment of the present application is schematically shown. Figure 4 As shown, the method includes the following processes:

[0099] Step 1: Before the intelligent agent operates, the system must first update the user's historical interactions and latest inputs in the information matching and mixing module, including knowledge information, historical interactions, historical tool selections, and thinking arrangement information, to form a comprehensive contextual view, thereby enhancing personalized services and contextual understanding capabilities.

[0100] Step 2: In the information mixing and matching module, firstly, different types of information are mixed and searched through vectorization and vector search enhancement, and more rich and relevant context text is generated by aggregating relevant information in different memory types (i.e., the above-mentioned knowledge information, historical interaction, historical tool selection and thinking arrangement information, which correspond to the previous knowledge data, historical interaction data, historical agent selection data and thinking arrangement data, respectively). Therefore, the information mixing and matching module can perform accurate and fast search based on the user's historical interaction and the latest input.

[0101] In order to improve the speed and accuracy of retrieval, the four memory types need to be first segmented into fixed-length text segments and then converted into vector representations. The vectorized text segments are stored in a vector database that supports vector-based search operations such as nearest neighbor search. The hybrid retrieval system includes embedded vector retrieval and an inverted index-based search engine. In order to capture the semantic intent of the query and retrieve information that is semantically similar to the query, embedded vector retrieval technology is introduced to find the most relevant document fragments to the query by measuring the similarity between vectors. In order to quickly locate documents containing specific keywords, this module uses an inverted index-based search engine to represent text segments as sparse vectors and quickly locate documents containing specific keywords through the inverted index. Since the module includes four different types of memory information and the types are complex, it is necessary to combine the vector-based search results with the keyword-based search results to form a unified retrieval result set through deduplication and integration.

[0102] At the same time, the information mixing and matching module updates the data in the memory in real time to ensure the timeliness and relevance of the information. This dynamicity enables the system to flexibly respond to changes in user needs, improving the system's adaptability and user satisfaction.

[0103] Step 3: Obtain information about time, space, and corresponding vertical fields through the information perception enhancement module to avoid the spread of outdated information and provide multi-dimensional information for task planning.

[0104] That is, to ensure that the information provided by the system is up to date and to avoid the spread of outdated information, although the large model can achieve a certain degree of information perception, it still has certain problems in terms of multi-dimensionality, accuracy, reliability and information perception in highly complex scenarios. Therefore, it is necessary to establish an information perception enhancement module to provide the system with multi-dimensional and accurate perception information. Specifically, the following perception enhancement processing can be provided:

[0105] Provide time-aware tools, including but not limited to date query, holiday query, weather query, and time difference query, so that during the task planning stage, the system can plan and execute tasks based on time information.

[0106] Provide spatial perception tools, including but not limited to geocoding, map services, spatial analysis tools, navigation and path planning tools, and GIS integration tools, etc. This enables the intelligent system to better understand and respond to user needs related to geographic location and provide richer and more personalized services.

[0107] In the scenario of vertical domain intelligent agents, perception information of the domain is often needed. Taking the network big model as an example, it is necessary to add network status perception, network topology perception, network security perception and network performance perception, so that the network big model intelligent agent can better understand and adapt to the network environment and provide more efficient, secure and personalized services.

[0108] Step 4: Using the collected information and combining it with the perception information obtained by information perception enhancement, the system constructs a comprehensive prompt to guide the large language model (LLMs) to generate a task plan. This prompt includes the agent configuration file, user-specified behavior guidelines, tool specifications, integrated memory text box, current timestamp, and predefined output format.

[0109] Step 5: The system makes tool calls based on the plan made by the agent-generated task plan. The results of tool execution in this stage are also integrated into the task memory so that the system can use this information in subsequent task planning to improve the continuity and consistency of decision-making.

[0110] Step 6: The system decides whether to complete the loop based on the tasks generated in the task planning phase. If not, it continues to execute the specified command, execute the tool, and re-enter the task planning phase based on the execution results and updated memory information until the end condition is met or the loop ends after reaching the predetermined maximum number of iterations.

[0111] Step 7: In this phase, the system forms a final response to the user's query based on the retrieved memory and the user's instructions. This process is similar to the "task planning" phase, but the focus is on generating the user's final answer.

[0112] The technical solution of this application can achieve the following effects:

[0113] 1. Provides a more coherent and efficient user experience: Through the information mixing and matching module, it can cache user context from multiple dimensions such as knowledge information and historical conversations, significantly improving the context understanding and memory capabilities of the intelligent agent. This enhanced capability makes the intelligent agent more accurate and personalized in the conversation.

[0114] 2. Success rate and efficiency of task execution: The information perception enhancement module enables the intelligent agent to comprehensively consider time, space and vertical field information when planning and executing tasks, improving the intelligent agent's thinking and decision-making ability. Compared with the existing technology, the technical solution of this application can more comprehensively understand and respond to complex task requirements.

[0115] 3. Improve the response speed and task processing efficiency of the intelligent agent: Based on the information mixing and matching and perception enhancement modules, the intelligent agent autonomous circulation system in the technical solution of this application improves the autonomy of the intelligent agent, reduces dependence on human intervention, and enables the intelligent agent to respond to changes and perform tasks more quickly and independently.

[0116] It should be noted that although the steps of the method in the present application are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0117] The following introduces an apparatus embodiment of the present application, which can be used to execute the information processing method in the above-mentioned embodiment of the present application. Figure 5 The structure block diagram of the information processing device provided by the embodiment of the present application is schematically shown. Figure 5 As shown, the information processing device provided in the embodiment of the present application includes:

[0118] Retrieval module 510, used to search in a preset database according to user input information, and obtain database retrieval information matching the user input information;

[0119] The enhancement module 520 is used to perform information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information;

[0120] A prompt generating module 530, configured to generate task prompt information according to the user input information, the database search information and the perception enhancement information;

[0121] The processing module 540 is used to generate target result information corresponding to the user input information according to the task prompt information.

[0122] In one embodiment of the present application, the retrieval module 510 includes:

[0123] A first retrieval unit, configured to perform vectorization processing on the user input information to obtain a user input vector, and search the preset database according to the user input vector to obtain first retrieval information;

[0124] A second retrieval unit, configured to extract keywords from the user input information, and search the preset database according to the keywords to obtain second retrieval information;

[0125] An integration unit is used to integrate the first search information and the second search information to generate database search information matching the user input information.

[0126] In one embodiment of the present application, the data types in the preset database include knowledge data, historical interaction data, historical agent selection data and thought arrangement data; the first retrieval unit is specifically used to:

[0127] Determine data of each data type that matches the user input vector according to the similarity between the user input vector and the data vectors corresponding to each data type in the preset database;

[0128] The first search information is obtained according to data of each data type that matches the user input vector.

[0129] In one embodiment of the present application, the second retrieval unit is specifically used to:

[0130] Obtain an inverted index corresponding to each data type in the preset database, wherein the inverted index indicates an index relationship between a data keyword and the data in which it is located;

[0131] Search the inverted index corresponding to each data type according to the keyword to determine the data corresponding to each data type that contains the keyword;

[0132] The second search information is obtained according to the data containing the keyword corresponding to each data type.

[0133] In one embodiment of the present application, the integration unit is specifically used for:

[0134] Deduplication processing is performed on data of each data type that matches the user input vector and data corresponding to each data type that contains the keyword to obtain database search information that matches the user input information.

[0135] In one embodiment of the present application, the enhancement module 520 is specifically used to:

[0136] Performing notification enhancement processing on the user input information in terms of time dimension, space dimension and vertical domain information dimension respectively, to obtain time enhancement information, space enhancement information and vertical domain enhancement information respectively;

[0137] At least one of the temporal enhancement information, the spatial enhancement information and the vertical field enhancement information is used as the perception enhancement information corresponding to the user input information.

[0138] In one embodiment of the present application, the processing module 540 is specifically used to:

[0139] Calling a designated agent to process the user input information according to the task prompt information to obtain agent output information;

[0140] When the output condition is not met, the agent output information is added to the preset database, and the process returns to the step of searching the preset database according to the user input information;

[0141] When the output condition is met, the agent output information is used as the target result information corresponding to the user input information; wherein, satisfying the output condition includes that the number of times the agent processes the user input information reaches a preset number of iterations, or the evaluation index corresponding to the agent output information meets the index requirement.

[0142] The specific details of the information processing device provided in each embodiment of the present application have been described in detail in the corresponding method embodiments and will not be repeated here.

[0143] Figure 6 The structure block diagram of a computer system for implementing an electronic device according to an embodiment of the present application is schematically shown.

[0144] It should be noted that Figure 6 The computer system 600 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0145] like Figure 6 As shown, the computer system 600 includes a central processing unit 601 (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory 602 (ROM) or the program loaded from the storage part 608 to the random access memory 603 (RAM). Various programs and data required for system operation are also stored in the random access memory 603. The central processing unit 601, the read-only memory 602 and the random access memory 603 are connected to each other through a bus 604. An input / output interface 605 (Input / Output interface, i.e., I / O interface) is also connected to the bus 604.

[0146] The following components are connected to the input / output interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a local area network card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read therefrom is installed into the storage section 608 as needed.

[0147] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processor 601, various functions defined in the system of the present application are executed.

[0148] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by an instruction execution system, device or device or used in combination with it. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0149] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0150] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0151] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0152] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application.

[0153] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. An information processing method, characterized in that: include: Searching in a preset database according to the user input information to obtain database search information matching the user input information; Performing information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information; generating task prompt information according to the user input information, the database search information and the perception enhancement information; Generate target result information corresponding to the user input information according to the task prompt information.

2. The information processing method according to claim 1, characterized in that: Searching in a preset database according to the user input information to obtain database search information matching the user input information includes: Performing vectorization processing on the user input information to obtain a user input vector, and searching in the preset database according to the user input vector to obtain first search information; Extracting keywords from the user input information, and searching the preset database according to the keywords to obtain second search information; The first search information and the second search information are integrated to generate database search information matching the user input information.

3. The information processing method according to claim 2, characterized in that: The data types in the preset database include knowledge data, historical interaction data, historical agent selection data and thought arrangement data; The first search information is obtained by searching the preset database according to the user input vector, including: Determine data of each data type that matches the user input vector according to the similarity between the user input vector and the data vectors corresponding to each data type in the preset database; The first search information is obtained according to data of each data type that matches the user input vector.

4. The information processing method according to claim 3, characterized in that: The preset database is searched according to the keyword to obtain second search information, including: Obtain an inverted index corresponding to each data type in the preset database, wherein the inverted index indicates an index relationship between a data keyword and the data in which it is located; Search the inverted index corresponding to each data type according to the keyword to determine the data corresponding to each data type that contains the keyword; The second search information is obtained according to the data containing the keyword corresponding to each data type.

5. The information processing method according to claim 4, characterized in that: Integrating the first search information and the second search information to generate database search information matching the user input information includes: Deduplication processing is performed on data of each data type that matches the user input vector and data corresponding to each data type that contains the keyword to obtain database search information that matches the user input information.

6. The information processing method according to claim 1, characterized in that: Performing information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information includes: Performing notification enhancement processing on the user input information in terms of time dimension, space dimension and vertical domain information dimension respectively, to obtain time enhancement information, space enhancement information and vertical domain enhancement information respectively; At least one of the temporal enhancement information, the spatial enhancement information and the vertical field enhancement information is used as the perception enhancement information corresponding to the user input information.

7. The information processing method according to claim 1, characterized in that: Generating target result information corresponding to the user input information according to the task prompt information includes: Calling a designated agent to process the user input information according to the task prompt information to obtain agent output information; When the output condition is not met, the agent output information is added to the preset database, and the process returns to the step of searching the preset database according to the user input information; When the output condition is met, the agent output information is used as the target result information corresponding to the user input information; wherein, satisfying the output condition includes that the number of times the agent processes the user input information reaches a preset number of iterations, or the evaluation index corresponding to the agent output information meets the index requirement.

8. An information processing device, characterized in that: include: A search module, used to search in a preset database according to user input information to obtain database search information matching the user input information; an enhancement module, configured to perform information perception enhancement processing on the user input information to obtain perception enhancement information corresponding to the user input information; a prompt generation module, configured to generate task prompt information according to the user input information, the database search information and the perception enhancement information; A processing module is used to generate target result information corresponding to the user input information according to the task prompt information.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the information processing method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor executes the executable instructions so that the electronic device executes the information processing method described in any one of claims 1 to 7.

Images