Method for processing requirements related to map points of interest, electronic device and storage medium

Abstract

The present disclosure provides a method and device for processing requirements related to map points of interest, electronic equipment and storage medium, which relates to the fields of artificial intelligence technologies such as generative large language models, agents, search-enhanced generation. The method comprises: obtaining dynamic requirements related to map points of interest, the dynamic requirements being requirements that need to be processed in combination with real-time information associated with map points of interest; issuing requirement parameters corresponding to the dynamic requirements to a target agent with corresponding requirement processing capability, the target agent being constructed in advance based on an agent technology combining map point of interest related services and search-enhanced generation technology; controlling the target agent to search for real-time information associated with the map points of interest to obtain supplementary information; and controlling the target agent to output a requirement processing result corresponding to the supplementary information and the requirement parameters as input information.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, in particular to the technical field of artificial intelligence such as generative large language models, agents, retrieval-augmented generation, and more particularly to a method and device for processing a demand related to a point of interest on a map, an electronic device, a computer readable storage medium, and a computer program product. BACKGROUND

[0002] Large language models (LLM, Large Language Model, which is essentially a generative model, also known as generative large language model) such as ChatGPT (Chat Generative Pre-trained Transformer, a chat robot program developed by OpenAI) can generate fluent responses similar to humans for many downstream tasks such as task-oriented dialogue and question answering.

[0003] However, how to better meet the user's timeliness demand (also known as dynamic demand, i.e., a demand that needs to combine real-time information to dynamically determine the result) related to a certain point of interest (POI) on a map is a key point that needs to be solved by those skilled in the art, for example, "How long does it take to drive from A company to B city before work next week?" and "What time do I need to leave home to catch the 10:00 am flight tomorrow?" Such demands not only need to combine real-time information but also need to fully understand the semantics. SUMMARY

[0004] The present disclosure provides a method and device for processing a demand related to a point of interest on a map, an electronic device, a computer readable storage medium, and a computer program product.

[0005] In a first aspect, the present disclosure provides a method for processing a demand related to a point of interest on a map, comprising: obtaining a dynamic demand related to a point of interest on a map; wherein the dynamic demand is a demand that needs to be processed in combination with real-time information associated with the point of interest on a map; issuing a demand parameter corresponding to the dynamic demand to a target agent having a corresponding demand processing capability; wherein the target agent is constructed in advance based on an agent technology that combines a point of interest related service and retrieval-augmented generation technology; controlling the target agent to retrieve real-time information associated with the point of interest on a map to obtain supplementary information; and controlling the target agent to output a demand processing result corresponding to the supplementary information and the demand parameter as input information.

[0006] In a second aspect, the embodiments of the present disclosure provide a device for processing a demand related to a map interest point, comprising: a dynamic demand obtaining unit configured to obtain a dynamic demand related to the map interest point; wherein the dynamic demand is a demand that needs to be processed in combination with real-time information associated with the map interest point; a demand parameter issuing unit configured to issue a demand parameter corresponding to the dynamic demand to a target agent having a corresponding demand processing capability; wherein the target agent is constructed in advance based on an agent technology combined with a map interest point related service and a search enhancement generation technology; a real-time information searching unit configured to control the target agent to search for real-time information associated with the map interest point to obtain supplementary information; and a demand processing result output unit configured to control the target agent to output a demand processing result corresponding to the supplementary information and the demand parameter as input information.

[0007] In a third aspect, the embodiments of the present disclosure provide an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the method for processing a demand related to a map interest point as described in the first aspect.

[0008] In a fourth aspect, the embodiments of the present disclosure provide a non-transitory computer readable storage medium storing computer instructions for enabling a computer to implement the method for processing a demand related to a map interest point as described in the first aspect.

[0009] In a fifth aspect, the embodiments of the present disclosure provide a computer program product comprising a computer program, which, when executed by a processor, enables the steps of the method for processing a demand related to a map interest point as described in the first aspect.

[0010] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] Other features, objects, and advantages of the present disclosure will become more apparent from the following detailed description of non-limiting embodiments thereof as taken in conjunction with the accompanying drawings:

[0012] Figure 1 is an exemplary system architecture to which the present disclosure can be applied;

[0013] Figure 2 is a flowchart of a method for processing a demand related to a map interest point provided by the embodiments of the present disclosure;

[0014] Figure 3 A column list diagram of various target intelligent agents provided by the embodiments of the present disclosure is shown.

[0015] Figure 4 A flowchart of another processing method of requirements related to map interest points provided by the embodiments of the present disclosure is shown.

[0016] Figure 5 A flowchart of a method of distributing task parameters of respective sub-tasks to corresponding target intelligent agents provided by the embodiments of the present disclosure is shown.

[0017] Figure 6 A flowchart of another method of distributing task parameters of respective sub-tasks to corresponding target intelligent agents provided by the embodiments of the present disclosure is shown.

[0018] Figure 7 A flowchart of a method of retrieving supplementary information according to task parameters for each target intelligent agent provided by the embodiments of the present disclosure is shown.

[0019] Figure 8 A structural block diagram of a processing device of requirements related to map interest points provided by the embodiments of the present disclosure is shown.

[0020] Figure 9 A structural schematic diagram of an electronic device suitable for executing a processing method of requirements related to map interest points provided by the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0021] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, and should be considered as merely exemplary. Thus, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description. It should be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0022] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solutions comply with relevant laws and regulations and do not violate public order and good customs.

[0023] Figure 1 An exemplary system architecture 100 of the embodiments of the processing method, device, electronic device and computer readable storage medium of requirements related to map interest points of the present disclosure is shown.

[0024] As Figure 1As shown, the system architecture 100 can include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0025] A user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 and the server 105 can be installed with various applications for realizing information communication between them, such as map applications, online shopping applications, instant messaging applications, etc.

[0026] The terminal devices 101, 102, 103 and the server 105 can be hardware or software. When the terminal devices 101, 102, 103 are hardware, they can be various electronic devices with display screens, including but not limited to smartphones, tablet computers, laptop computers, desktop computers, etc. When the terminal devices 101, 102, 103 are software, they can be installed in the above-mentioned electronic devices, and can be implemented as multiple software or software modules, or as a single software or software module, which is not specifically limited here. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server 105 is software, it can be implemented as multiple software or software modules, or as a single software or software module, which is not specifically limited here.

[0027] The server 105 can provide various services through various built-in applications. Taking a map application that can process dynamic requirements related to map points of interest initiated by a user as an example, the server 105 can achieve the following effects when running the map application: first, receiving a dynamic requirement related to a map point of interest transmitted by a user through a terminal device 101, 102, 103 through the network 104, the dynamic requirement being a requirement that needs to be processed in combination with real-time information associated with the map point of interest; then, issuing a requirement parameter corresponding to the dynamic requirement to a target agent with corresponding requirement processing capability, the target agent being pre-built based on an agent technology combining map point of interest related services and search enhancement generation technology; next, controlling the target agent to search for real-time information associated with the map point of interest to obtain supplementary information; finally, controlling the target agent to output a requirement processing result corresponding to the supplementary information and the requirement parameter as input information.

[0028] Furthermore, the server 105 can also return the result of the request processing to the terminal devices 101, 102, and 103 via the network 104, so that the terminal devices 101, 102, and 103 can present the received request processing result to the user.

[0029] It should be noted that dynamic requests related to map points of interest can be temporarily obtained from terminal devices 101, 102, and 103 via network 104, or they can be pre-stored locally on server 105 through various means. Therefore, when server 105 detects that this data is already stored locally (e.g., when it starts processing previously reserved dynamic requests), it can choose to directly obtain this data from locally. In this case, the exemplary system architecture 100 may not include terminal devices 101, 102, and 103 and network 104.

[0030] The processing methods for map point of interest (POI) related requirements provided in the subsequent embodiments of this disclosure are generally executed by a server 105 with strong computing power and abundant computing resources. Correspondingly, the processing device for map POI related requirements is also generally located in the server 105. However, it should also be noted that when terminal devices 101, 102, and 103 also have sufficient computing power and resources, they can also complete the aforementioned calculations performed by the server 105 through map applications installed on them, and thus output the same results as the server 105. Especially when multiple terminal devices with different computing capabilities exist simultaneously, but the map application determines that the terminal device has strong computing power and abundant remaining computing resources, it can allow the terminal device to perform the aforementioned calculations, thereby appropriately reducing the computing pressure on the server 105. Accordingly, the processing device for map POI related requirements can also be located in the terminal devices 101, 102, and 103. In this case, the exemplary system architecture 100 may also exclude the server 105 and the network 104.

[0031] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0032] Please refer to Figure 2 , Figure 2 A flowchart of a method for processing map point of interest related requirements provided in this disclosure embodiment, wherein process 200 includes the following steps:

[0033] Step 201: Obtain dynamic requirements related to points of interest on the map;

[0034] This step is intended for the implementation of methods for handling needs related to map points of interest (e.g.,Figure 1 The server 105 or the map application installed and running on the terminal device 101, 102, 103 obtains a dynamic demand of a user for a map application initiated in relation to a map point of interest, which refers to a demand that can be processed only based on real-time change information of the map point of interest at a specific time point based on a certain demand scenario. That is, such a demand cannot be accurately processed only by static and fixed map knowledge or map information, but is affected by various dynamic factors such as time, environment, external conditions, etc. In the application scenario of maps and navigation, dynamic demand often requires real-time or near real-time feedback and processing. The dynamic demand can be a question, demand or expectation described by the user in various forms of natural language.

[0035] For example, the user asks "How long does it take to drive from A company to B city next Monday?", which needs to consider various dynamic factors such as real-time traffic, weather, traffic control, road restrictions, congestion, etc. in addition to solving the basic problem of path planning. These factors change with changes in time, location, etc.

[0036] Among them, the dynamic demand related to the map point of interest refers to the actual demand expressed by the user, which at least contains a fuzzy or determined map point of interest (such as a company, a city, a tourist attraction, etc.), which can be used as a retrieval positioning object, or as a starting point or end point in navigation parameters or path planning parameters.

[0037] At the same time, the processing of dynamic demand is not only a simple calculation about time and place, but also needs to be calculated according to real-time traffic conditions, weather, holidays, etc. For example, the traffic conditions in a certain time period (such as road closure, construction, etc.), and the road condition prediction data related to the time period (such as congestion, accidents, number limit information, etc.), all of which belong to real-time data.

[0038] Still taking the dynamic demand "time from A company to B city next Monday" as an example, it not only involves the geographical position of the starting point and the end point and the preset path, but also needs to be dynamically evaluated according to the traffic data of the future time (next Monday). The acquisition of this real-time data can include but is not limited to:

[0039] 1) Traffic flow data: including real-time road traffic conditions, congestion prediction;

[0040] 2) Traffic control information: such as number limit, road closure, construction section, etc.;

[0041] 3) Weather impact: such as the impact of rainfall, snowstorm, etc. on road traffic;

[0042] 4) Other emergencies: such as temporary events caused by traffic accidents, road accidents, etc.

[0043] An implementation includes, but is not limited to:

[0044] Firstly, the execution subject obtains an arbitrary demand related to a map interest point; then, when the execution subject identifies that the arbitrary demand contains a preset keyword (for example, specific time description information such as "next Monday" and "tomorrow", or prediction-related description information such as "estimated time consumption") representing a demand processing requirement combined with real-time information associated with the map interest point, the execution subject can determine the arbitrary demand as the dynamic demand.

[0045] Step 202: issuing a demand parameter corresponding to the dynamic demand to a target agent with corresponding demand processing capability;

[0046] On the basis of step 201, this step aims to issue, by the execution subject, a demand parameter corresponding to the dynamic demand to a target agent with corresponding demand processing capability. The target agent is pre-constructed based on an agent technology combining a map interest point related service and retrieval-augmented generation (RAG) technology. Specifically, the target agent can include at least one of the following (see the schematic diagram shown in Figure 3

[0047] a retrieval positioning agent corresponding to a retrieval positioning service of a map interest point; a path planning agent corresponding to a path planning service between different map interest points; a navigation agent corresponding to a navigation service between different map interest points; and an associated recommendation agent corresponding to an associated recommendation service of a map interest point.

[0048] The agent pre-constructed in this step is based on various services or functions provided by a map application for a map interest point. Not only does it use an agent technology to make it have stronger semantic understanding and complex problem processing capability, but it also combines RAG technology to improve the accuracy and knowledge richness of the generative large language model used when forming the agent when processing information, especially enabling it to obtain real-time information from the outside to solve dynamic demands. The core idea of the RAG technology is to obtain information related to user queries (in this disclosure, the user initiates a dynamic demand related to a map interest point) through an external retrieval system, and combine these information with other capabilities of the generative large language model, thereby generating answers with more context awareness and knowledge richness.

[0049] The technical principle of RAG includes two main steps:

[0050] ​Retrieval: Upon receiving the user's query or input, the agent constructed by combining the RAG technique can first retrieve relevant information related to the input content (e.g., map points of interest) from an external document database, knowledge base, or search engine through a retriever (usually a semantic search-based tool such as an inverted index or neural network retrieval);

[0051] Generation: The retrieved relevant information is input into a generative large language model (such as GPT series, BERT, etc.) along with the user's original input as context. This information serves as supplementary information to help the generative large language model understand the original input better, enabling it to generate more accurate, informative, and accurate text content as an answer.

[0052] Step 203: Control the target agent to retrieve real-time information associated with the map point of interest to obtain supplementary information.

[0053] This step aims to control the target agent to retrieve real-time information associated with the map point of interest to obtain supplementary information by the above-mentioned execution subject. That is, the retrieval phase of the agent combined with the RAG technique when processing dynamic requirements.

[0054] Step 204: Control the target agent to output the demand processing result corresponding to the input information of the supplementary information and the demand parameter.

[0055] Based on step 203, this step aims to control the target agent to output the demand processing result corresponding to the input information of the supplementary information and the demand parameter by the above-mentioned execution subject. That is, the generation phase of the agent combined with the RAG technique when processing dynamic requirements.

[0056] The method for processing requirements related to map points of interest provided by the embodiments of the present disclosure is aimed at dynamic requirements initiated by users that require real-time information associated with the map point of interest for demand processing. By constructing an agent based on the combination of map point of interest-related services and retrieval enhancement generation technology in advance, the demand parameter corresponding to the dynamic requirement can be issued to a target agent with corresponding demand processing capabilities. The target agent can obtain real-time information associated with the map point of interest as supplementary information through retrieval, and ultimately make the target agent output a demand processing result that better meets the user's actual demand corresponding to the input information of the supplementary information and the demand parameter.

[0057] Considering that the dynamic requirement expressed by the user to the map application may be a complex requirement that cannot be completely processed through a single-step operation, please refer to Figure 4 , Figure 4Another flowchart of a method for processing a demand related to a map interest point is provided in the embodiments of the present disclosure, wherein the flowchart 400 includes the following steps:

[0058] Step 401: Obtain a dynamic demand related to a map interest point.

[0059] This step is consistent with step 201 in flowchart 200, which will not be repeated here.

[0060] Step 402: Use a preset generative large language model to split the dynamic demand into at least two sub-tasks.

[0061] This step aims to use a pre-trained generative large language model to perform task decomposition on the dynamic demand, so as to split it into at least two sub-tasks. All sub-tasks are executed to complete the processing of the actual dynamic demand.

[0062] In the process of splitting the user's dynamic demand into multiple sub-tasks, the generative large language model often needs to first fully understand the semantics to obtain the expected processing intent, and then perform task splitting according to the expected processing intent and the preset problem processing logic to obtain the sub-tasks corresponding to each processing step.

[0063] An implementation mode, including but not limited to, can be:

[0064] Using a generative large language model to split the dynamic demand into at least two sub-tasks in a thought chain mode.

[0065] The generative large language model is a large-scale natural language processing model trained through deep learning technology. This type of model can generate corresponding output according to input text and is commonly used for tasks such as dialogue generation, automatic text writing, and translation. The generative large language model can process demands by understanding and generating text and produce appropriate responses based on context. In this embodiment, it is used to implement task decomposition for complex user demands.

[0066] The thought chain mode (Chain of Thought, COT) refers to the decomposition of complex problems into a series of logically ordered thinking steps. This method is commonly used to solve tasks that require multiple reasoning steps. The generative large language model decomposes the dynamic demand into multiple sub-tasks through the "thought chain mode" and reasons through the logical order of each sub-task. For example, when processing a problem, a preliminary analysis may be performed first, followed by the derivation of necessary information, and finally a conclusion is drawn or a specific operation is performed. These steps form a chain, and each sub-task is the reasoning result of the previous task. Through this mode, complex problems can be simplified into a series of manageable small tasks, which can be solved step by step.

[0067] Step 403: determining the target agent corresponding to each subtask;

[0068] On the basis of step 402, this step aims to determine the target agent corresponding to each subtask by the above-mentioned execution subject. Specifically, the process of assigning each subtask to the matching target agent can also be carried out by the generative large language model, i.e. the generative large language model needs to register each constructed agent in advance, and make it determine the matching target agent for each split subtask according to the pre-learned correspondence between different subtasks and different agents. Of course, in addition to the above operation by the map application or the generative large language model controlled by the map application, another agent specialized in subtask matching can also be used to perform the operation, which is not limited here as long as the matching target agent for each subtask can be determined.

[0069] Step 404: issuing the task parameters corresponding to each subtask to the corresponding target agent;

[0070] On the basis of step 403, this step aims to issue the task parameters corresponding to each subtask to the corresponding target agent by the above-mentioned execution subject. Among them, the task parameters contain the task information required by the target agent to process the corresponding subtask, for example, assuming that there is a subtask to determine the planned travel route between A place and B place, the task parameters should at least contain the POI information of A place as the starting point and the POI information of B place as the destination, as well as the selected limit parameters containing possible travel time, travel mode and travel preferences, etc.

[0071] Among them, the task parameters of each subtask can also be the task description information corresponding to each subtask generated by the generative large language model, and when determining the target agent matching each subtask, the target agent matching each subtask can also be determined according to the task description information corresponding to each subtask.

[0072] Step 405: controlling the target agent to retrieve real-time information associated with map points of interest to obtain supplementary information;

[0073] Step 406: controlling the target agent to output the demand processing result corresponding to the supplementary information and the demand parameter as input information.

[0074] The above steps 405-406 are consistent with steps 203-204 as shown in Figure 2 , and the same part of the content please refer to the corresponding part of the above embodiment, which will not be repeated here.

[0075] Compared with Figure 2The embodiment shown herein is implemented through... Figure 4 The provided implementation leverages a generative large language model in steps 4021 and 403 to break down dynamic requirements into multiple sub-tasks and identify the target agent corresponding to each sub-task. This facilitates the distribution of task parameters for each sub-task to the corresponding target agent. This enhances the ability to handle complex dynamic requirements and, through collaboration between different target agents corresponding to different sub-tasks, better addresses user dynamic requirements, resulting in more user-expected and evidence-based requirements processing outcomes.

[0076] To deepen the understanding of how to determine the matching target agent for each subtask and how to distribute task parameters to the corresponding target agent, it is also disclosed that... Figure 5 and Figure 6 Two different implementation methods are provided, among which Figure 5 The illustrated process 500 includes the following steps:

[0077] Step 501: Obtain the task description information generated by the generative large language model for each subtask;

[0078] Step 502: Extract the target function name from the task description information corresponding to each subtask;

[0079] This step aims to have the aforementioned executing entity extract the target function name from the task description information corresponding to each subtask, meaning that the task description information must at least contain the function name and task parameters of the function capable of handling the subtask.

[0080] For example, if the subtask is a path planning task, then the function name recorded in its corresponding task description information can be: path planning function.

[0081] Step 503: Based on the preset function registry, determine the registered intelligent agent with the same target function name as the target intelligent agent corresponding to the subtask;

[0082] The function registry records the correspondence between different function names and different registered agents. Pre-built agents with different function names are registered with the aforementioned execution entity, thus forming the function registry. This step aims to allow the execution entity to determine, based on the correspondence recorded in the function registry, the registered agents with the same target function name as the target agents for the corresponding subtasks.

[0083] Step 504: Send the task parameters corresponding to each subtask to the corresponding target agent.

[0084] The step is to issue the task parameters extracted from the task description information corresponding to each subtask to the corresponding target agent by the above-mentioned subject.

[0085] Figure 6 The flow 600 shown comprises the following steps:

[0086] Step 601: obtaining the task description information respectively generated by the generative large language model for each subtask;

[0087] Step 602: extracting the target function name from the task description information corresponding to each subtask;

[0088] Steps 601-602 are consistent with steps 501-502, which will not be repeated here.

[0089] Step 603: determining the target routing address corresponding to the target function name according to the preset function routing registration table;

[0090] The function routing registration table records the correspondence between different function names and different routing addresses, and different routing addresses are used to call different registered agents. The agent with different function names constructed in advance will be registered in the above-mentioned execution subject, thereby forming the function routing registration table.

[0091] The step is to determine the target routing address corresponding to the target function name according to the correspondence recorded in the function routing registration table by the above-mentioned execution subject.

[0092] Step 604: determining the registered agent that can be called through the target routing address as the target agent corresponding to the corresponding subtask;

[0093] On the basis of step 603, the step is to determine the registered agent that can be called through the target routing address as the target agent corresponding to the corresponding subtask by the above-mentioned execution subject.

[0094] Step 605: extracting the corresponding task parameters from the task description information corresponding to each subtask;

[0095] Step 606: passing the extracted task parameters to the corresponding target agent through the corresponding target routing address.

[0096] Steps 605-606 are to pass the task parameters extracted from the task description information to the corresponding target agent through the target routing address by the above-mentioned execution subject, so as to realize the task parameter issuing to the target agent.

[0097] In Figure 4The solution provided by the illustrated embodiment is based on the fact that the complex dynamic requirements are divided into multiple sub-tasks by means of the generative large language model, so that the process 203 will also adapt to the changes: the above-mentioned execution subject respectively controls each target agent to search the real-time information of the map interest points contained in the received task parameters, obtains the search results, and the supplementary information includes the search results.

[0098] To deepen the understanding of how the target agent searches the real-time information of the map interest points contained in the task parameters, please refer to Figure 7 , Figure 7 A flowchart of a method for searching for supplementary information according to task parameters for each target agent is provided by the embodiment of the present disclosure, which aims to control each target agent to perform the following search steps as shown in flowchart 700:

[0099] Step 701: Extracting map interest points from received task parameters;

[0100] Step 702: Calling a preset search tool to search a search statement constructed by the map interest points as search keywords, and obtaining real-time search results returned by the preset search tool;

[0101] Among them, the preset search tool can include at least one of the preset search engine, the preset geographic knowledge graph, the preset road information source, and the preset lane-level road database.

[0102] Among them, the map interest point refers to a specific geographic location or object on the map, which is of interest to the user, such as restaurants, stores, scenic spots, hospitals, etc. Taking it as a search keyword means taking it as a keyword or search item, which may be the name, classification, description or other identification information related to the location. And the constructed search statement can be a complete search statement or query string constructed using the related information of the map interest point, which can be a search request for a specific field. For example, the search statement can be "find nearby restaurants", or "query medical facilities in the area", etc.

[0103] And the construction of the search statement depends on the input interest point, and may combine specific geographic location, time or other filtering conditions. The key of this step is to dynamically generate a query according to the interest point data, rather than a static query, which can combine the user's historical interaction data with the map application or the user's real-time portrait or preferences, etc.

[0104] Step 703: Generating supplementary prompt words according to real-time search results.

[0105] On the basis of step 702, the present step aims to generate a supplementary prompt word by the above-mentioned execution body according to the real-time retrieval result. The supplementary prompt word can form more comprehensive input information together with the original dynamic demand expression or the task parameter of the current subtask, so as to output more accurate answers by the target agent to the input information.

[0106] On the basis of how to control the target agent to retrieve the supplementary information respectively given in the above embodiment, the demand processing result matched with the dynamic demand can also be obtained by the following way:

[0107] The target agent corresponding to the last subtask is controlled by the above-mentioned execution body to output the demand processing result corresponding to the input information composed of the corresponding supplementary information and the task parameter.

[0108] Among them, the last subtask is the subtask that is executed last in the execution sequence of each subtask, and the execution sequence is determined according to the execution dependency relationship between each subtask (which is usually reflected after the subtasks are split in the way of thinking chain), and the task parameter of the subtask executed later is determined based on the subtask processing result of the subtask executed earlier. That is, the input information of the target agent of the last subtask has collected the output result of the target agent of the previous subtask, so the output result of the target agent of the last subtask can be taken as the demand processing result of the dynamic demand.

[0109] In order to deepen the understanding, the present disclosure also gives a specific implementation scheme in combination with a specific application scenario:

[0110] Suppose the dynamic demand input by the user is to determine the actual time consumption from the first map interest point to the second map interest point in a preset time period, then the dynamic demand can be first split into sequentially executed retrieval positioning subtask, path planning subtask and navigation subtask in the way of thinking chain by using the generative large language model.

[0111] Among them, the retrieval positioning subtask is used to locate the actual geographical positions of the first map interest point and the second map interest point, the path planning subtask is used to determine the travel route according to the actual geographical positions of the first map interest point and the second map interest point, and the navigation subtask is used to determine the actual time consumption of traveling in the travel route in the preset time period.

[0112] After the retrieval positioning subtask is issued to the retrieval positioning agent, the retrieval positioning agent can be controlled to retrieve the first real-time information associated with the interest point names of the first map interest point and the second map interest point, and obtain the first supplementary information;

[0113] After the path planning subtask is assigned to the path planning agent, the path planning agent can be controlled to retrieve second real-time information associated with each actual road segment covered by the travel route and the preset time period, to obtain second supplementary information;

[0114] After the navigation subtask is assigned to the navigation agent, the navigation agent can be controlled to retrieve third real-time information associated with travel on the travel route by different vehicles in the preset time period, to obtain third supplementary information.

[0115] Finally, the navigation agent is controlled to output a predicted time consumption corresponding to the third supplementary information and the corresponding task parameters as input information.

[0116] The task parameters assigned to the navigation agent are determined based on the task parameters of the navigation subtask and the subtask processing result of the path planning subtask, the subtask processing result of the path planning subtask is determined based on the task parameters of the path planning subtask, the second supplementary information, and the subtask processing result of the retrieval positioning subtask, and the subtask processing result of the retrieval positioning subtask is determined based on the task parameters of the retrieval positioning subtask and the first supplementary information.

[0117] Still taking "how long does it take to drive from A company to B city next Monday" as an example, the execution subject of the embodiment can first split it into a POI retrieval task and assign it to the retrieval positioning agent to obtain the geographic location information of the starting point (A company) and the ending point (B city) of the user;

[0118] Then the path planning task is further split and assigned to the path planning agent, the path planning agent obtains the geographic location information of A company and B city from the retrieval positioning agent, queries multiple candidate travel routes and travel road segments between the two locations, and optimizes them in combination with real-time traffic information, weather information, and license number restriction information in the future time (next Monday) to obtain an optimized travel route with as little congestion as possible;

[0119] Finally, the navigation task is further split and assigned to the navigation agent, the navigation agent obtains the optimized travel route from the path planning agent, gives navigation information in combination with the selected travel tool, and extracts the predicted travel time consumption from the navigation information and feeds it back to the user.

[0120] Further reference Figure 8 , as an implementation of the method shown in the above figures, the disclosure provides an embodiment of a device for processing requirements related to map points of interest, which corresponds to the method embodiment shown in Figure 2 The device can be applied to various electronic devices.

[0121] As Figure 8As shown, the processing apparatus 800 for the demand related to the map interest point in the embodiment can include: a dynamic demand obtaining unit 801, a demand parameter issuing unit 802, a real-time information retrieving unit 803, and a demand processing result output unit 804. The dynamic demand obtaining unit 801 is configured to obtain a dynamic demand related to the map interest point. The dynamic demand is a demand that needs to be processed in combination with real-time information associated with the map interest point. The demand parameter issuing unit 802 is configured to issue a demand parameter corresponding to the dynamic demand to a target agent having a corresponding demand processing capability. The target agent is pre-built based on an agent technology combining a map interest point related service and a search enhancement generation technology. The real-time information retrieving unit 803 is configured to control the target agent to search for real-time information associated with the map interest point to obtain supplementary information. The demand processing result output unit 804 is configured to control the target agent to output a demand processing result corresponding to the supplementary information and the demand parameter as input information.

[0122] In the embodiment, the dynamic demand obtaining unit 801, the demand parameter issuing unit 802, the real-time information retrieving unit 803, and the demand processing result output unit 804 in the processing apparatus 800 for the demand related to the map interest point can respectively refer to the specific processing and the technical effects brought by the same in the corresponding embodiment. Figure 2 The related description of steps 201-204 in the corresponding embodiment will not be repeated here.

[0123] In some optional implementation manners of the embodiment, the dynamic demand obtaining unit 801 can be further configured to:

[0124] obtain any demand related to the map interest point;

[0125] in response to the any demand containing a preset keyword representing a need to process the demand in combination with real-time information associated with the map interest point, determine that the obtained any demand is the dynamic demand.

[0126] In some optional implementation manners of the embodiment, the target agent includes at least one of the following:

[0127] a search positioning agent corresponding to a search positioning service of the map interest point;

[0128] a path planning agent corresponding to a path planning service between different map interest points;

[0129] a navigation agent corresponding to a navigation service between different map interest points;

[0130] an association recommendation agent corresponding to an association recommendation service of the map interest point.

[0131] In some optional implementations of the embodiment, the processing apparatus 800 for handling requirements related to map interest points can further include:

[0132] a requirement splitting unit configured to split the dynamic requirement into at least two subtasks by using a preset generative large language model;

[0133] a matching agent determination unit configured to determine a target agent corresponding to each subtask;

[0134] Correspondingly, the requirement parameter issuing unit 802 can include:

[0135] a task parameter issuing subunit configured to issue a task parameter corresponding to each subtask to the corresponding target agent.

[0136] In some optional implementations of the embodiment, the requirement splitting unit can include:

[0137] a thought chain mode splitting subunit configured to split the dynamic requirement into at least two subtasks in a thought chain mode by using a generative large language model.

[0138] In some optional implementations of the embodiment, the thought chain mode splitting subunit can be further configured to:

[0139] in response to the dynamic requirement being to determine an actual time consumption from a first map interest point to a second map interest point within a preset time period, split the dynamic requirement into sequentially executed search and positioning subtasks, path planning subtasks, and navigation subtasks in a thought chain mode by using a generative large language model; wherein the search and positioning subtasks are used to locate actual geographic positions of the first map interest point and the second map interest point, the path planning subtasks are used to determine a travel route according to the actual geographic positions of the first map interest point and the second map interest point, and the navigation subtasks are used to determine an actual time consumption of traveling along the travel route within the preset time period.

[0140] In some optional implementations of the embodiment, the matching agent determination unit can include:

[0141] a task description information acquisition subunit configured to acquire task description information respectively generated by the generative large language model for each subtask;

[0142] a matching agent determination subunit configured to determine a target agent corresponding to each subtask according to the task description information corresponding to each subtask.

[0143] In some optional implementations of the embodiment, the matching agent determination subunit can include:

[0144] The function name extraction module is configured to extract the target function name from the task description information corresponding to each subtask;

[0145] The agent matching module is configured to determine, according to a preset function registry, a registered agent with the same target function name as the target agent corresponding to the respective subtask; wherein the function registry records the correspondence between different function names and different registered agents.

[0146] In some optional implementations of the embodiment, the agent matching module can be further configured to:

[0147] In response to the function registry being a function routing registry, determine, according to the function routing registry, a target routing address corresponding to the target function name; wherein the function routing registry records the correspondence between different function names and different routing addresses, and different routing addresses are used to call different registered agents;

[0148] The registered agent that can be called through the target routing address is determined as the target agent corresponding to the respective subtask.

[0149] In some optional implementations of the embodiment, the task parameter issuing subunit can be further configured to:

[0150] Extract the respective task parameters from the task description information corresponding to each subtask, respectively;

[0151] Pass the extracted task parameters to the respective target agents through the respective target routing addresses, respectively.

[0152] In some optional implementations of the embodiment, the real-time information retrieval unit 803 can include:

[0153] The real-time information retrieval subunit is configured to control each target agent to perform real-time information retrieval on the map interest points contained in the received task parameters, respectively, to obtain retrieval results; wherein the supplementary information includes the retrieval results.

[0154] In some optional implementations of the embodiment, the real-time information retrieval subunit can be further configured to:

[0155] For each target agent, control the target agent to perform the following retrieval steps:

[0156] Extract the map interest points from the received task parameters;

[0157] Call a preset retrieval tool to perform retrieval on a retrieval statement constructed from the map interest points serving as retrieval keywords, to obtain real-time retrieval results returned by the preset retrieval tool;

[0158] generate a supplementary prompt word according to the real-time search result; wherein the supplementary information comprises the supplementary prompt word.

[0159] In some optional implementations of the embodiment, the preset search tool comprises any one of the following:

[0160] The preset search engine, the preset geographic knowledge graph, the preset road information source, and the preset lane-level road database.

[0161] In some optional implementations of the embodiment, the demand processing result output unit 804 can be further configured to:

[0162] control the target agent corresponding to the last subtask to output a demand processing result corresponding to the respective supplementary information and the task parameter as input information; wherein the last subtask is a subtask that is executed last in execution time sequence, and the execution time sequence is determined according to the execution dependency relationship between the subtasks, and the task parameter of the subtask executed later is determined based on the subtask processing result of the subtask executed earlier.

[0163] In some optional implementations of the embodiment, the real-time information search unit 803 can comprise a further configuration to:

[0164] in response to the search positioning subtask being issued to the search positioning agent, control the search positioning agent to search for first real-time information associated with the interest point names of the first map interest point and the second map interest point, to obtain first supplementary information;

[0165] in response to the path planning subtask being issued to the path planning agent, control the path planning agent to search for second real-time information associated with each actual road segment covered by the travel route and the preset time period, to obtain second supplementary information;

[0166] in response to the navigation subtask being issued to the navigation agent, control the navigation agent to search for third real-time information associated with traveling along the travel route by different means of transportation in the preset time period, to obtain third supplementary information.

[0167] In some optional implementations of the embodiment, the demand processing result output unit 804 can be further configured to:

[0168] The control navigation intelligent agent outputs a predicted time consumption corresponding to the third supplementary information and the corresponding task parameter as input information; wherein, the task parameter issued to the navigation intelligent agent is determined based on the task parameter of the navigation subtask and the subtask processing result of the path planning subtask, the subtask processing result of the path planning subtask is determined based on the task parameter of the path planning subtask, the second supplementary information and the subtask processing result of the search and positioning subtask, and the subtask processing result of the search and positioning subtask is determined based on the task parameter of the search and positioning subtask and the first supplementary information.

[0169] The embodiment provided in the embodiment corresponds to the above-mentioned method embodiment, and the embodiment provides a processing device for a demand related to a map interest point. For a dynamic demand initiated by a user and requiring processing of real-time information associated with the map interest point, an intelligent agent is constructed based on an intelligent agent technology combined with a map interest point related service and a search enhancement generation technology. The demand parameter corresponding to the dynamic demand is issued to a target intelligent agent having a corresponding demand processing capability, so that the target intelligent agent can obtain the real-time information associated with the map interest point as supplementary information through searching, and finally make the target intelligent agent output a demand processing result more in line with the actual demand of the user corresponding to the supplementary information and the demand parameter as input information.

[0170] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the processing method of the demand related to the map interest point described in any of the above embodiments.

[0171] According to an embodiment of the present disclosure, the present disclosure further provides a readable storage medium storing computer instructions for enabling a computer to implement the processing method of the demand related to the map interest point described in any of the above embodiments.

[0172] According to an embodiment of the present disclosure, the present disclosure further provides a computer program product, which, when executed by a processor, can implement the processing method of the demand related to the map interest point described in any of the above embodiments.

[0173] Figure 9A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0174] As shown in Figure 9 The device 900 includes a computing unit 901 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 902 or a computer program loaded into a random access memory (RAM) 903 from a storage unit 908. Various programs and data required for the operation of the device 900 can also be stored in the RAM 903. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0175] Various components in the device 900 are connected to the I / O interface 905, including an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, a magneto-optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0176] The computing unit 901 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs various methods and processes described above, such as the processing method for map point of interest related demand. For example, in some embodiments, the processing method for map point of interest related demand can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded onto the RAM 903 and executed by the computing unit 901, one or more steps of the processing method for map point of interest related demand described above can be performed. Alternatively, in other embodiments, the computing unit 901 can be configured to perform the processing method for map point of interest related demand by any other suitable means, such as by means of firmware.

[0177] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0178] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a function / operation specified in the flowchart and / or block diagram. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0179] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0180] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0181] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0182] The computer system can include clients and servers. This relationship can be

[0183] According to the technical scheme of the embodiments of the present disclosure, for the dynamic demand initiated by the user and requiring the demand processing combined with the real-time information associated with the map interest point, the agent is constructed in advance based on the agent technology combined with the search enhancement generation technology, so as to issue the demand parameter corresponding to the dynamic demand to the target agent with the corresponding demand processing capability, and then the target agent can obtain the real-time information associated with the map interest point as supplementary information through the search mode, and then the target agent outputs the demand processing result corresponding to the supplementary information and the demand parameter as the input information, which is more in line with the actual demand of the user.

[0184] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present disclosure can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical scheme of the present disclosure can be achieved, which is not limited herein.

[0185] The above detailed description does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for processing requirements related to map points of interest, comprising: Obtain any request related to a map point of interest; in response to the fact that the arbitrary request contains a preset keyword that indicates the need to process the request in conjunction with real-time information associated with the map point of interest, determine that the obtained arbitrary request is a dynamic request. The demand parameters corresponding to the dynamic demand are sent to the target intelligent agent with the corresponding demand processing capability. The target intelligent agent is pre-built based on the intelligent agent technology that combines map point of interest related services and retrieval enhancement generation technology. The target intelligent agent is controlled to retrieve real-time information associated with the map points of interest to obtain supplementary information; The target intelligent agent is controlled to output a demand processing result corresponding to the supplementary information and the demand parameters as input information.

2. The method according to claim 1, wherein, The target intelligent agent includes at least one of the following: A search and location agent corresponding to the map point of interest search and location service; A path planning agent corresponding to the path planning service between different map points of interest; Navigation agents corresponding to navigation services between different map points of interest; The association recommendation agent corresponding to the association recommendation service with map points of interest.

3. The method according to claim 2, further comprising: Using a pre-defined generative large language model, the dynamic requirements are broken down into at least two sub-tasks; Identify the target agent corresponding to each of the sub-tasks; Correspondingly, the step of sending the demand parameters corresponding to the dynamic demand to the target intelligent agent with the corresponding demand processing capability includes: The task parameters corresponding to each subtask are sent to the corresponding target agent.

4. The method according to claim 3, wherein, The process utilizes a pre-defined generative large language model to break down the dynamic requirements into at least two sub-tasks, including: Using the generative large language model, the dynamic requirements are broken down into at least two sub-tasks in a thought chain pattern.

5. The method according to claim 4, wherein, The generative large language model is used to break down the dynamic requirements into at least two sub-tasks using a thought chain pattern, including: In response to the dynamic demand for determining the actual time taken to travel from the first map point of interest to the second map point of interest within a preset time period, the generative large language model is used to break down the dynamic demand into sequentially executed sub-tasks of retrieval and location, route planning, and navigation, according to the thought chain approach. The retrieval and location sub-task is used to locate the actual geographical locations of the first and second map points of interest; the route planning sub-task is used to determine the travel route based on the actual geographical locations of the first and second map points of interest; and the navigation sub-task is used to determine the actual time taken to travel according to the travel route within the preset time period.

6. The method according to claim 3, wherein, The determination of the target agent corresponding to each of the sub-tasks includes: Obtain the task description information generated by the generative large language model for each of the subtasks; Based on the task description information corresponding to each subtask, the target intelligent agent corresponding to the corresponding subtask is determined.

7. The method according to claim 6, wherein, The step of determining the target agent corresponding to each subtask based on the task description information corresponding to each subtask includes: Extract the target function name from the task description information corresponding to each of the subtasks; According to the preset function registry, registered agents with the same target function name are identified as the target agents corresponding to the subtasks; wherein, the function registry records the correspondence between different function names and different registered agents.

8. The method according to claim 7, wherein, The step of determining, according to a preset function registry, the registered agent with the same target function name as the target agent for the corresponding subtask includes: In response to the fact that the function registry is a function routing registry, the target routing address corresponding to the target function name is determined according to the function routing registry; wherein, the function routing registry records the correspondence between different function names and different routing addresses, and different routing addresses are used to call different registered smart agents; Registered agents that can be invoked through the target routing address are identified as the target agents corresponding to the respective subtasks.

9. The method according to claim 8, wherein, The step of sending the task parameters corresponding to each subtask to the corresponding target agent includes: Extract the corresponding task parameters from the task description information corresponding to each of the sub-tasks; The extracted task parameters are transmitted to the corresponding target agent through the corresponding target routing address.

10. The method according to claim 3, wherein, The control of the target agent retrieves real-time information associated with the map points of interest to obtain supplementary information, including: Each target agent is controlled to retrieve map points of interest contained in the received task parameters in real time to obtain retrieval results; wherein, the supplementary information includes the retrieval results.

11. The method according to claim 10, wherein, The step of controlling each target agent to retrieve map points of interest contained in the received task parameters in real time, and obtaining retrieval results, includes: For each target agent, the following retrieval steps are executed: Extract the map points of interest from the received task parameters; The preset search tool is invoked to search for the search statement constructed using the map points of interest as search keywords, and the real-time search results returned by the preset search tool are obtained. Supplementary suggestions are generated based on the real-time search results; wherein, the supplementary information includes the supplementary suggestions.

12. The method according to claim 11, wherein, The preset search tool includes any one of the following: Preset search engine, preset geographic knowledge graph, preset road administration information source, preset lane-level road database.

13. The method according to claim 10, wherein, The control of the target agent to output a demand processing result corresponding to the supplementary information and the demand parameters as input information includes: The target agent corresponding to the final subtask outputs the corresponding requirement processing result, which is based on the supplementary information and task parameters as input information. The final subtask is the subtask that is executed last in the execution sequence among all the subtasks. The execution sequence is determined according to the execution dependency relationship between the subtasks. The task parameters of the subtask executed later are determined based on the subtask processing result of the subtask executed earlier.

14. The method according to claim 5, wherein, The control of the target agent retrieves real-time information associated with the map points of interest to obtain supplementary information, including: In response to the retrieval and location subtask being sent to the retrieval and location agent, the retrieval and location agent is controlled to retrieve first real-time information associated with the names of the first map point of interest and the second map point of interest, and obtain first supplementary information. In response to the route planning subtask being sent to the route planning agent, the route planning agent is controlled to retrieve second real-time information associated with each actual road segment covered by the travel route and the preset time period, thereby obtaining second supplementary information; In response to the navigation subtask being sent to the navigation agent, the navigation agent is controlled to retrieve third real-time information related to travel using different modes of transportation along the travel route during the preset time period, thereby obtaining third supplementary information.

15. The method according to claim 14, wherein, The control of the target agent to output a demand processing result corresponding to the supplementary information and the demand parameters as input information includes: The navigation agent is controlled to output a predicted time corresponding to the input information of the third supplementary information and the corresponding task parameters; wherein, the task parameters issued to the navigation agent are determined based on the task parameters of the navigation subtask and the subtask processing results of the path planning subtask, the subtask processing results of the path planning subtask are determined based on the task parameters of the path planning subtask, the second supplementary information and the subtask processing results of the retrieval and positioning subtask, and the subtask processing results of the retrieval and positioning subtask are determined based on the task parameters of the retrieval and positioning subtask and the first supplementary information.

16. A processing apparatus for map point of interest related requirements, comprising: The dynamic demand acquisition unit is configured to acquire any demand related to map points of interest. In response to the fact that any demand contains preset keywords that represent the need to process the demand in conjunction with real-time information associated with the map points of interest, the obtained demand is determined to be a dynamic demand. The demand parameter distribution unit is configured to distribute the demand parameters corresponding to the dynamic demand to the target intelligent agent with the corresponding demand processing capability; wherein, the target intelligent agent is pre-built based on intelligent agent technology that combines map point of interest related services and retrieval enhancement generation technology; The real-time information retrieval unit is configured to control the target intelligent agent to retrieve real-time information associated with the map points of interest and obtain supplementary information; The demand processing result output unit is configured to control the target agent to output a demand processing result corresponding to the supplementary information and the demand parameters as input information.

17. The apparatus according to claim 16, wherein, The target intelligent agent includes at least one of the following: A search and location agent corresponding to the map point of interest search and location service; A path planning agent corresponding to the path planning service between different map points of interest; Navigation agents corresponding to navigation services between different map points of interest; The association recommendation agent corresponding to the association recommendation service with map points of interest.

18. The apparatus of claim 17, further comprising: The requirement splitting unit is configured to split the dynamic requirement into at least two sub-tasks using a preset generative large language model; The matching agent determination unit is configured to determine the target agent corresponding to each of the sub-tasks; Correspondingly, the requirement parameter distribution unit includes: The task parameter distribution subunit is configured to distribute the task parameters corresponding to each subtask to the corresponding target agent.

19. The apparatus according to claim 18, wherein, The demand splitting unit includes: The thought chain pattern is configured to use the generative large language model to break down the dynamic requirements into at least two sub-tasks using the thought chain pattern.

20. The apparatus according to claim 19, wherein, The sub-units of the mind chain model are further configured as follows: In response to the dynamic demand for determining the actual time taken to travel from the first map point of interest to the second map point of interest within a preset time period, the generative large language model is used to break down the dynamic demand into sequentially executed sub-tasks of retrieval and location, route planning, and navigation, according to the thought chain approach. The retrieval and location sub-task is used to locate the actual geographical locations of the first and second map points of interest; the route planning sub-task is used to determine the travel route based on the actual geographical locations of the first and second map points of interest; and the navigation sub-task is used to determine the actual time taken to travel according to the travel route within the preset time period.

21. The apparatus according to claim 18, wherein, The matching agent determination unit includes: The task description information acquisition subunit is configured to acquire task description information generated by the generative large language model for each of the subtasks. The matching agent determination subunit is configured to determine the target agent corresponding to the corresponding subtask based on the task description information corresponding to each of the subtasks.

22. The apparatus according to claim 21, wherein, The matching agent determination subunit includes: The function name extraction module is configured to extract the target function name from the task description information corresponding to each of the subtasks; The agent matching module is configured to determine, based on a preset function registry, registered agents with the same target function name as the target agents for the corresponding subtasks; wherein, the function registry records the correspondence between different function names and different registered agents.

23. The apparatus according to claim 22, wherein, The agent matching module is further configured to: In response to the fact that the function registry is a function routing registry, the target routing address corresponding to the target function name is determined according to the function routing registry; wherein, the function routing registry records the correspondence between different function names and different routing addresses, and different routing addresses are used to call different registered smart agents; Registered agents that can be invoked through the target routing address are identified as the target agents corresponding to the respective subtasks.

24. The apparatus according to claim 23, wherein, The task parameter distribution subunit is further configured to: Extract the corresponding task parameters from the task description information corresponding to each of the sub-tasks; The extracted task parameters are transmitted to the corresponding target agent through the corresponding target routing address.

25. The apparatus according to claim 18, wherein, The real-time information retrieval unit includes: The real-time information retrieval subunit is configured to control each of the target agents to retrieve real-time information on the map points of interest contained in the received task parameters, and obtain retrieval results; wherein, the supplementary information includes the retrieval results.

26. The apparatus according to claim 25, wherein, The real-time information retrieval subunit is further configured to: For each target agent, the following retrieval steps are executed: Extract the map points of interest from the received task parameters; The preset search tool is invoked to search for the search statement constructed using the map points of interest as search keywords, and the real-time search results returned by the preset search tool are obtained. Supplementary suggestions are generated based on the real-time search results; wherein, the supplementary information includes the supplementary suggestions.

27. The apparatus according to claim 26, wherein, The preset search tool includes any one of the following: Preset search engine, preset geographic knowledge graph, preset road administration information source, preset lane-level road database.

28. The apparatus according to claim 25, wherein, The demand processing result output unit is further configured to: The target agent corresponding to the control and the last subtask outputs the required processing result corresponding to the supplementary information and task parameters as input information; wherein, the last subtask is the subtask that is executed last in the execution sequence among all the subtasks, the execution sequence is determined according to the execution dependency relationship between the subtasks, and the task parameters of the subtask executed later are determined based on the subtask processing result of the subtask executed earlier.

29. The apparatus according to claim 20, wherein, The real-time information retrieval unit includes components further configured to: In response to the retrieval and location subtask being sent to the retrieval and location agent, the retrieval and location agent is controlled to retrieve first real-time information associated with the names of the first map point of interest and the second map point of interest, and obtain first supplementary information. In response to the route planning subtask being sent to the route planning agent, the route planning agent is controlled to retrieve second real-time information associated with each actual road segment covered by the travel route and the preset time period, thereby obtaining second supplementary information; In response to the navigation subtask being sent to the navigation agent, the navigation agent is controlled to retrieve third real-time information related to travel using different modes of transportation along the travel route during the preset time period, thereby obtaining third supplementary information.

30. The apparatus according to claim 29, wherein, The demand processing result output unit is further configured to: The navigation agent is controlled to output a predicted time corresponding to the input information of the third supplementary information and the corresponding task parameters; wherein, the task parameters issued to the navigation agent are determined based on the task parameters of the navigation subtask and the subtask processing results of the path planning subtask, the subtask processing results of the path planning subtask are determined based on the task parameters of the path planning subtask, the second supplementary information and the subtask processing results of the retrieval and positioning subtask, and the subtask processing results of the retrieval and positioning subtask are determined based on the task parameters of the retrieval and positioning subtask and the first supplementary information.

31. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the processing method for map point of interest related requirements as described in any one of claims 1-15.

32. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the processing method for map point of interest related requirements as described in any one of claims 1-15.

33. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the processing method for map point of interest related requirements according to any one of claims 1-15.

Images