Information acquisition method and device, storage medium and electronic equipment

Abstract

The invention discloses an information collection method and device, a storage medium and electronic equipment. The method comprises the steps that an information collection task is acquired, and a collection data element set corresponding to the information collection task is generated; determining an information collection role corresponding to each data element in the collection data element set, and configuring and generating a collection session request; the information acquisition role comprises information input personnel and / or information acquisition equipment; sending a collection session request to the information collection role, generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model, and obtaining data parameters corresponding to the collection data element set based on the collection session; and encoding the data parameter into an information acquisition result in a preset format. According to the invention, the technical problems of low information acquisition accuracy, low efficiency and poor information acquisition flexibility due to the fact that information acquisition, processing and application are solidified in a vertical field in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an information collection method and device, a storage medium and an electronic device. Background Art

[0002] During the use of the information collection system in the related technology, the collection, processing and application of information are solidified in a vertical field. When sharing and applying data between different vertical fields, the architecture of this information collection system has problems such as inconsistent data accuracy, granularity, specifications, and quality, as well as low information collection efficiency and repeated reporting by personnel. Moreover, information entry personnel often cannot flexibly increase or decrease information items or information volume according to differences in information collection scenarios, nor can they coordinate work with other information entry personnel and automatic collection systems, which limits the elastic expansion of information collection tasks. Summary of the invention

[0003] The embodiments of the present application provide an information collection method and device, a storage medium and an electronic device to at least solve the technical problem that the collection, processing and application of information in the related art are solidified in a vertical field, resulting in low accuracy, low efficiency and poor flexibility of information collection.

[0004] According to one aspect of an embodiment of the present application, there is provided an information collection method, including: acquiring an information collection task, generating a collection data element set corresponding to the information collection task; determining an information collection role corresponding to each data element in the collection data element set, and configuring and generating a collection session request; the information collection role includes an information entry person and / or an information collection device; sending a collection session request to the information collection role, generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model, and acquiring data parameters corresponding to the collection data element set based on the collected session; encoding the data parameters into an information collection result in a preset format, and determining that the information collection task is completed.

[0005] According to another aspect of an embodiment of the present application, an information collection device is also provided, including: an acquisition generation unit, used to acquire an information collection task, and generate a collection data element set corresponding to the information collection task; a determination configuration unit, used to determine the information collection role corresponding to each data element in the collection data element set, and configure to generate a collection session request; the information collection role includes an information entry personnel and / or an information collection device; a generation unit, used to send a collection session request to the information collection role, and generate a collection session corresponding to the information collection task based on the collection session request and a preset large language model; an acquisition unit, used to acquire data parameters corresponding to the collection data element set based on the collected session; a first determination unit, used to encode the data parameters into an information collection result in a preset format, and determine that the information collection task is completed.

[0006] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the information collection method through the computer program.

[0007] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned information collection method when running.

[0008] The embodiments of the present application have the following beneficial technical effects:

[0009] 1. This application can break down the barriers between information collection and application in different vertical fields, realize timely, accurate, complete and standard information collection, avoid problems such as repeated collection, improve the quality and efficiency of existing manual information collection work, and reduce the data usage cost of the application system.

[0010] Second, through active information collection process scheduling, the coordinated work of multiple data entry personnel and automated collection equipment can be achieved, which not only ensures the quality and efficiency of information collection tasks, avoids repeated collection, but also significantly improves the flexibility of information collection tasks.

[0011] 3. This application supports voice interaction, multimodal information collection, automated equipment collection, and proactive collection process management, which can significantly reduce the workload of manual information collection and improve user experience.

[0012] Fourth, based on the large language model, not only can the standardized extraction of information be completed, but also the information collected throughout the entire life cycle can be sorted, summarized, and managed on the timeline, which greatly enriches the scenarios of data application. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0014] Figure 1 is a schematic diagram of an application environment of an optional information collection method according to an embodiment of the present application;

[0015] Figure 2 is a schematic diagram of an application environment of another optional information collection method according to an embodiment of the present application;

[0016] Figure 3 It is a schematic diagram of an information collection process of an optional information collection system in the related art;

[0017] Figure 4 is a flowchart of an optional information collection method according to an embodiment of the present application;

[0018] Figure 5 is a schematic diagram of an information collection process of an optional information collection system according to an embodiment of the present application;

[0019] Figure 6 is a schematic diagram of the architecture of an optional information collection system according to an embodiment of the present application;

[0020] Figure 7 is a schematic structural diagram of an optional information collection device according to an embodiment of the present application;

[0021] Figure 8 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] According to one aspect of the embodiments of the present application, an information collection method is provided. Optionally, as an optional implementation, the information collection method can be but is not limited to being applied to: Figure 1 In the application environment shown. The application environment includes: a terminal device 102 for human-computer interaction with a user, a network 104, and a server 106. A user 108 can perform human-computer interaction with the terminal device 102, and an information collection client runs in the terminal device 102. The terminal device 102 includes a human-computer interaction screen 1022, a processor 1024, and a memory 1026. The human-computer interaction screen 1022 is used to display information collection tasks and information collection results; the processor 1024 is used to obtain information collection tasks. The memory 1026 is used to store and display information collection tasks and information collection results.

[0025] In addition, the server 106 includes a database 1062 and a processing engine 1064, and the database 1062 is used to store information collection tasks and information collection results. The processing engine 1064 is used to obtain information collection tasks, generate a collection data element set corresponding to the information collection task; determine the information collection role corresponding to each data element in the collection data element set, and configure and generate a collection session request; the information collection role includes information entry personnel and / or information collection equipment; send a collection session request to the information collection role, generate a collection session corresponding to the information collection task based on the collection session request and a preset large language model, and obtain data parameters corresponding to the collection data element set based on the collection session; encode the data parameters into an information collection result in a preset format, determine that the information collection task is completed; and send the above information collection result to the client of the above terminal device 102.

[0026] In one or more embodiments, the information collection method of the present application can be applied to Figure 2 In the application environment shown. Figure 2As shown, human-computer interaction can be performed between user 202 and user device 204. User device 204 includes memory 206 and processor 208. In this embodiment, user device 204 can refer to but not be limited to executing the operations performed by the above-mentioned terminal device 102 to display the information collection results of the information collection task.

[0027] Optionally, the terminal device 102 and the user device 204 include but are not limited to mobile phones, tablet computers, laptop computers, PCs, vehicle-mounted electronic devices, wearable devices and other terminals, and the network 104 may include but is not limited to wireless networks or wired networks. Among them, the wireless network includes: WIFI and other networks that realize wireless communication. The wired network may include but is not limited to: wide area network, metropolitan area network, local area network. The server 106 may include but is not limited to any hardware device that can perform computing. The server may be a single server, or a server cluster composed of multiple servers, or a cloud server. The above is only an example, and no limitation is made to this in this embodiment.

[0028] like Figure 3 As shown in the figure, during the use of the current mainstream manual information collection system, the collection, processing and application of information are solidified in a vertical field. When sharing and applying data between different vertical fields, the architecture of this information collection system has problems such as inconsistent data accuracy, granularity, specifications, and quality, as well as low information collection efficiency and repeated reporting by personnel. Moreover, information entry personnel often cannot flexibly increase or decrease information items or information volume according to differences in information collection scenarios, nor can they coordinate work with other information entry personnel and automatic collection systems, which limits the elastic expansion of information collection tasks.

[0029] To solve the above technical problems, as an optional implementation method, Figure 4 As shown, the embodiment of the present application provides an information collection method, comprising the following steps:

[0030] S402, obtaining an information collection task, and generating a collection data element set corresponding to the information collection task;

[0031] S404, determining the information collection role corresponding to each data element in the collection data element set, and configuring and generating a collection session request; the information collection role includes an information entry person and / or an information collection device;

[0032] S406, sending a collection session request to the information collection role, and generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model;

[0033] S408, obtaining data parameters corresponding to the collected data element set based on the collected session;

[0034] S410, encoding the data parameters into an information collection result in a preset format, and determining that the information collection task is completed.

[0035] Specifically, in the embodiments of the present application, Figure 5 and Figure 6 As shown, the interactive information collection system of the embodiment of the present application is composed of a multimodal large model in the background, a scheduling module, a configuration database (DataBase, DB), a session DB, a unified collection front end in the foreground (a session interface for interacting with the input personnel), and an application programming interface (Application Programming Interface, API) of the automatic collection device. The scheduling module receives the collection requirements sent by the application system, calls the multimodal large model to parse the collection requirement parameters, and configures the generated collection session request in combination with the session generation parameter information, input personnel information, automatic collection device API configuration and other information stored in the configuration DB, and adds the personnel and equipment related to the information collection to the collection session. After starting the collection session, the input personnel and the automatic collection device actively or passively interact with the scheduling module and the multimodal large model through the session interface of the unified collection front end. After completing the information collection process together, the scheduling module returns the generated collection results to the application system.

[0036] It should be noted that an information collection task can be an information collection task with a complete collection life cycle. A collection life cycle has at least a start time and an end time of the task. Each time corresponds to an independent information collection task, which can be recorded as a first information collection task. Each first information collection task corresponds to a round of collection session.

[0037] In one implementation, the information collection task may be an information collection task at a specified time, that is, a single round of information collection task.

[0038] The above-mentioned collection task refers to the scheduling module generating the data element content specified by the collection task through a complete collection session process, through interaction with the input personnel and automatic collection equipment, according to the time or frequency specified by the collection task, and the collection requirements of the data elements of each collection node. The above-mentioned collection task includes the collection task identification, the life cycle of information collection, and the collection element set corresponding to the collection task.

[0039] The collection lifecycle includes the start and end times of a task, or the start or end conditions or instructions.

[0040] Data elements define the requirements for collecting each data element included in the collection task, including data specifications, collection frequency, quality checks, collection roles, etc. Among them, data specifications define the type of data, threshold range, etc.; collection frequency defines the timing of collection of the data element item during the collection life cycle; quality checks are used to determine whether the data is compliant and logical, and the information collection role can be the role of the data entry personnel or the automatic collection device. Furthermore, multiple collection roles can be set for the same data element, and different roles can back up each other to improve the collection success rate of data elements.

[0041] The embodiment of the present application improves the efficiency and accuracy of information collection through the above-mentioned automated and intelligent methods, while reducing manual intervention. By combining a large language model with automated equipment, the system can flexibly respond to different information collection needs, and can effectively interact with input personnel and automatic collection equipment, and can flexibly increase or decrease information items or information volume according to differences in information collection scenarios.

[0042] In one or more embodiments, obtaining an information collection task includes:

[0043] Acquire a first type of information collection task, where the first type of information collection task is a collection task preset by the system, and the preset collection task has fixed information collection parameters; the information collection parameters include at least one of an identification, a life cycle, and a collection data element set of the information collection task;

[0044] Alternatively, a second type of information collection task is obtained, where the second type of information collection task is a task generated by the large language model based on a collection task instruction described in text.

[0045] Specifically, in one embodiment of the present application, two types of collection tasks are accepted through the scheduling module: one type is the collection task called by the standard API agreed with the application system (the first type of information collection task). Since this type of task has standardized parameter definitions, it can directly extract the parameters such as the identification, life cycle, and collection element item set of the collection task. The second type of information collection task is a task generated by a text-described collection task instruction, which is sent by the scheduling module to the large model for parsing, extracts the collection task identification and parameter settings, and combines the standard definition of the collection task in the configuration DB to set and adjust the collection life cycle, increase or decrease the collection element items, and adjust the collection element parameters to generate a collection element set. In this application, the scheduling module can flexibly handle preset standardized tasks and customized tasks based on user descriptions, thereby improving the efficiency of information collection.

[0046] Specifically, when the information collection task is only an information collection requirement at a certain moment, the information collection parameters may include an identifier of the information collection task and a set of collected data elements.

[0047] In one or more embodiments, determining the information collection role corresponding to each data element in the collection data element set includes:

[0048] Determine the input personnel corresponding to the first data element in the collection data element set according to the correspondence information between the input personnel and the information collection role preset in the configuration information, and the working time information of the input personnel;

[0049] According to the second data element in the set of collected data elements collected by the automatic collection device indicated in the configuration information, an application interface of the automatic collection device corresponding to the second data element is determined.

[0050] Specifically, in the embodiment of the present application, first, the system refers to the corresponding information of the input personnel and the information collection roles preset in the configuration information, that is, the system has defined which input personnel are responsible for which types of information collection tasks or data elements. In addition, when determining the input personnel, the system will also consider their working time information to ensure that the assigned tasks can be completed within their working hours. Based on the above information, the system will determine which input personnel are responsible for collecting the first data element in the data element set, that is, the target data that needs to be manually entered.

[0051] Next, the system checks the configuration information for indications that data elements are collected by automatic collection devices. For data elements that need to be collected automatically, the system determines the APIs of the automatic collection devices that correspond to these data elements. These APIs allow the automatic collection devices to communicate with other parts of the system to transmit the collected data. The system then determines which automatic collection devices are responsible for collecting the second data element in the set of data elements, that is, the data that can be collected in an automated manner.

[0052] In one or more embodiments, generating a collection session corresponding to an information collection task based on a collection session request and a preset large language model includes:

[0053] For any information collection moment in the information collection task, the first data element and the collection requirement of the first data element at any information collection moment are sent to the large language model, the large language model is called to generate an information collection instruction for the information entry personnel, and the information collection instruction is pushed to the information entry personnel;

[0054] The second data element and the collection requirement of the second data element at any information collection moment are sent to the large language model, and the large language model is called to generate a call instruction for the automatic collection device application interface. It can be understood that the large language model supports multimodal data recognition, such as voice data, text data, video data, etc. In one embodiment of the present application, the large language model is trained based on LLM.

[0055] In one embodiment of the present application, the system determines any information collection moment in the life cycle of the information collection task, that is, determines when a specific data element needs to be collected, and then sends the first data element (data that needs to be manually entered) and its collection requirements at the information collection moment to the large language model. The system calls the large language model, generates information collection instructions for information entry personnel based on the sent data elements and requirements, and then pushes the generated information collection instructions to the information entry personnel so that the information entry personnel can collect data according to these instructions.

[0056] The system sends the second data element (data that can be collected by the automatic collection device) and its collection requirements at the same information collection time to the large language model. The system calls the large language model, generates a calling interface corresponding to the application interface of the automatic collection device based on the sent data elements and requirements, and then uses the generated calling interface to interact with the automatic collection device to trigger the automatic collection device to collect information.

[0057] In one embodiment of the present application, information collection is completed through a unified collection front end, and collection of session instructions and information entry by information entry personnel are completed through a unified information collection front end.

[0058] The above-mentioned information collection method based on the large language model not only improves the flexibility and efficiency of information collection, enabling the system to adapt to various complex information collection scenarios, but also ensures the accuracy and timeliness of information collection, while reducing manual intervention and improving the automation level of the information collection process.

[0059] In one or more embodiments, a collection session request is sent to an information collection role, and a collection session corresponding to the information collection task is generated based on the collection session request and a preset large language model; and data parameters corresponding to the collection data element set are obtained based on the collection session, including:

[0060] Sending a first collection session request of the current round to the information collection role, and generating a first collection session corresponding to the information collection task of the current round based on the first collection session request and a preset large language model;

[0061] Based on the first collection session, receiving first feedback information from the information collection role from the first collection session;

[0062] Extracting data parameters corresponding to the current round in the collection data element set from the first feedback information;

[0063] When the extracted data parameters do not meet the information collection task of the current round, a second collection session request is sent to the information collection role, and second feedback information fed back by the information collection role is obtained based on the second collection session request; and data parameters corresponding to the current round in the collection data element set are extracted from the second feedback information;

[0064] Repeat the above steps until the extracted data parameters meet the information collection requirements of the current round.

[0065] It can be understood that the information collection task of the current round is also the information collection task at the current moment.

[0066] Specifically, in the embodiment of the present application, the complete steps of a round of information collection process include: first, based on the information collection instruction, generate a collection session corresponding to the information collection task, send the current round of collection session request to the information collection role, then receive the feedback information of the collection role, and extract the data parameters corresponding to the collection data element set from the collection session; when the data parameters corresponding to the extracted collection data element set do not meet the information collection requirements, repeat the above steps, and repeat the cycle until the extracted data parameters meet the information collection requirements of the current round. It can be understood that a round of collection session corresponds to an information collection task at a time node. To complete an information collection task at a time point, one, two or even multiple interactive sessions are required. Each time an interactive session is generated, the large model needs to extract data parameters from the collection session to determine whether the collection data elements corresponding to the collection task are met. If not, the session instruction generation, session request configuration, and session interaction are repeated until the extracted data parameters meet the information collection requirements of the current round.

[0067] In one or more embodiments, based on the first collection session, receiving first feedback information of the information collection role from the first collection session; extracting data parameters corresponding to the current round in the collection data element set from the first feedback information includes:

[0068] receiving first feedback information from a data entry person or an automatic collection device from a first collection session;

[0069] Sending data processing information to the large language model; the data processing information includes the first feedback information and the data element items corresponding to the current round and the collection requirements of the data elements;

[0070] Based on the data processing information, the large language model is called to extract the corresponding data parameters of the collected data element set.

[0071] Specifically, in the embodiment of the present application, the system receives feedback information provided by the input personnel or the automatic collection device from the collection session. The above feedback information may include the collected data, the collection status, etc. The system combines the feedback information with the data element items corresponding to the information collection time and the collection requirements of the data elements to form data processing information and send it to the large language model, so that the large language model extracts the data parameters corresponding to the collection data element set corresponding to the collection time according to the feedback information.

[0072] In one embodiment of the present application, based on the data processing information, the large language model is called to extract the data parameters corresponding to the current round in the collection data element set, including:

[0073] Obtain the historical session data corresponding to the current round of collection sessions, as well as the historical data parameters extracted from each historical session;

[0074] The large language model is called to extract the corresponding data parameters of the current round in the collected data element set from the historical session data, historical data parameters and data processing information.

[0075] Specifically, in the above implementation, one round of collection sessions includes at least two interactive sessions. When extracting collection parameters from the collection sessions, it is necessary to extract all historical (interactive) sessions of the current round of collection sessions and extract the collection parameters together with the current interactive session, that is, to improve the accuracy of large language model collection parameter extraction by constructing context information.

[0076] In one or more embodiments, when the information task includes multiple rounds of information collection, the multiple rounds of collection sessions can be performed on the same collection front end, and based on the data processing information, the large language model is called to extract the data parameters corresponding to the current round in the collection data element set, including:

[0077] Obtain the historical session data corresponding to the collection session of the historical round, as well as the historical data parameters extracted in each historical round;

[0078] The large language model is called to extract the corresponding data parameters of the current round in the collected data element set from the historical session data, historical data parameters and data processing information.

[0079] Specifically, in the embodiment of the present application, the data extraction process of the current round needs to send the conversation data of the previous rounds, the information extraction results (if the data volume is large, the large model extraction summary is called instead), and the data of multiple sources and multiple conversations of the current round as input to the large language model to improve the accuracy of the information extraction of the large model. For a single data element item, the large language model only extracts a unique result.

[0080] In one or more embodiments, the information collection method further includes:

[0081] Perform quality checks on feedback information based on the preset correlation logic between different data items and the background knowledge base;

[0082] In the case where there are missed or illogical data items in the feedback information, it is determined that the extracted data parameters do not meet the information collection requirements of the current round.

[0083] Specifically, in the embodiment of the present application, the scheduling module analyzes the data items in the feedback information according to the preset correlation logic and background knowledge base between different data items. For example, it identifies the relationship between the data items and whether they conform to the known logic and background knowledge; based on the analysis results of the large language model, the scheduling module adjusts the illogical data items, such as correcting errors, filling missing values ​​or smoothing outliers, to ensure the accuracy and consistency of the data. After adjustment, the large language model extracts the corresponding data parameters of the collection data element set corresponding to the collection time, and completes the information collection task.

[0084] In one or more embodiments, the information collection method further includes:

[0085] According to the preset correlation logic between different data items and the background knowledge base, a context environment is constructed based on the feedback information to determine the data items that are missed or illogical in the feedback information; the feedback information includes at least one of the first feedback information and the second feedback information;

[0086] Based on reasoning, supplement or correct missed or illogical data items.

[0087] Specifically, in the embodiment of the present application, the scheduling module of the system builds a context environment based on the above feedback information according to the preset correlation logic between different data items and the background knowledge base. After building the context environment, the scheduling module can analyze the feedback information and identify which data items are missing and which data are wrong based on the in-depth understanding of the correlation logic of the data items and the application of background knowledge. The scheduling module uses its reasoning ability to supplement or correct the missed or illogical data items based on the existing data and information in the background knowledge base.

[0088] In one or more embodiments, encoding the data parameters into information collection results in a preset format includes:

[0089] Encode the data parameters into information collection results in XML or JSON format.

[0090] Specifically, the embodiment of the present application converts the data parameters into information collection results in a string format of XML or JSON, and then returns the information collection results to the application system.

[0091] In an application embodiment, in combination Figure 6 As shown, the above-mentioned information collection method includes:

[0092] In the first step, the scheduling module receives and parses the collection tasks sent by the application system and generates a set of collection data elements.

[0093] The scheduling module can accept two types of collection tasks: one is the collection task called by the standard API agreed with the application system. Since this type of task has standardized parameter definitions, it can directly extract the parameters such as the identification, life cycle, and collection element item set of the collection task. In order to support the scalability of collection tasks, the second type of collection task is the collection task instruction described in text, which is sent by the scheduling module to the large model (multimodal large model) for parsing and generates the collection task identification and parameter settings. Combined with the standard definition of the collection task in the configuration DB, it is set to adjust the collection life cycle, increase or decrease the collection element items, and adjust the collection element item parameters to generate a collection element item set.

[0094] The second step is to extract the elements related to the collection session and configure and generate the collection session request.

[0095] On the basis of clarifying the collection requirements of collection elements, extract the collection role of each data element.

[0096] The input personnel corresponding to each data element item is determined according to the corresponding information of input personnel and collection roles preset in the configuration DB, the on-duty information of input personnel, etc. If the data element item is collected by an automatic collection device, the API of the automatic collection device is mapped to the data element item.

[0097] The scheduling module sends a session joining request to the data entry personnel and automatic collection devices determined to participate in this collection task to generate a collection session.

[0098] The third step is to start a conversation and complete a round of information collection.

[0099] 3.1 For a given information collection moment, the scheduling module sorts out the data elements and requirements that need to be collected at the information collection moment, and classifies them according to the input personnel and automatic collection equipment:

[0100] For the data entry personnel, the data elements and requirements that need to be collected at that moment are sent to the big model, and the big model is called to generate information collection instructions for the data entry personnel, which are stored in the reply DB. At the same time, voice instructions are pushed to the data entry personnel on the unified collection front end.

[0101] For automatic collection devices, the data elements and requirements that need to be collected at that moment are sent to the big model, and the big model is called to generate a calling interface for the API of the automatic collection device, which is stored in the session DB and displayed on the unified collection front end.

[0102] 3.2 The dispatch system monitors the feedback from the input personnel and automatic collection equipment during the conversation.

[0103] If no feedback is received within the specified time, the instruction will be resent to the data entry personnel or automatic data collection equipment.

[0104] If feedback is received from the data entry personnel or automatic collection equipment, the feedback content is decoded and sent to the big model together with the data elements and requirements that need to be collected at that moment, and the big model is called to extract the data elements; note that this extraction process requires the previous rounds of conversation data, information extraction results (if the amount of data is large, the big model is called to extract the summary instead), and the data from multiple sources and multiple conversations in this round to be sent to the big model as input after decoding to improve the accuracy of the big model information extraction. For a single data element, the big model only extracts a unique result.

[0105] In addition, the scheduling module will evaluate the quality of the data element items according to the quality inspection method of the data element items. Furthermore, the correlation logic and background knowledge base between different data items can be preset to check the correlation logic relationship between different data items and between data items and background knowledge bases, and determine the data items that meet the specifications but do not meet the logic.

[0106] Furthermore, the correlation logic and background knowledge base between different data items can be preset, and a context environment can be built based on the collected data items to perform collaborative reasoning and intelligent supplementation and correction on missed or illogical data items.

[0107] 3.3 The scheduling system will sort out the items that were missed or collected with low quality in the previous round, and then proceed to step 3.1. If the information collection in this round meets the requirements, it will proceed to the fourth step.

[0108] It should be noted that in subsequent multiple rounds of conversations, for each data entry person, the big model needs to input the reference collection instructions of the previous round when generating new collection instructions to ensure the logic of instruction generation and complete the collection in as few conversation rounds as possible.

[0109] The fourth step is to generate standard information collection results and return them to the application system.

[0110] The scheduling system encodes the data elements extracted by the large model into information collection results in a standard format (XML or JSON) and returns them to the application system. If the scheduling system detects that the life cycle of the collection task has ended, it exits the current collection process; otherwise, it starts a new round of information collection and goes to the third step above.

[0111] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0112] According to another aspect of the embodiments of the present application, an information collection device for implementing the above-mentioned information collection method is also provided. Figure 7 As shown, the device comprises:

[0113] An acquisition generating unit 702 is used to acquire an information collection task and generate a collection data element set corresponding to the information collection task;

[0114] A determination configuration unit 704 is used to determine the information collection role corresponding to each data element in the collection data element set, and configure and generate a collection session request; the information collection role includes an information entry person and / or an information collection device;

[0115] The generating unit 706 is used to send a collection session request to the information collection role, and generate a collection session corresponding to the information collection task based on the collection session request and a preset large language model.

[0116] An acquisition unit 708, configured to acquire data parameters corresponding to a collection data element set based on the collected session;

[0117] The first determining unit 710 is used to encode the data parameters into an information collection result in a preset format, and determine that the information collection task is completed.

[0118] In one or more embodiments, obtaining the generation unit 702 includes:

[0119] The first acquisition module is used to acquire a first type of information collection task. The first type of information collection task is a collection task preset by the system. The preset collection task has fixed information collection parameters. The information collection parameters include at least one of the information collection task identification, life cycle and collection data element set.

[0120] 1; or, a second acquisition module, used to acquire a second type of information collection task, where the second type of information collection task is a task generated by a large language model based on a collection task instruction described in text.

[0121] In one or more embodiments, determining the configuration unit 704 includes:

[0122] A first determination module, used to determine the input personnel corresponding to the first data element in the collection data element set according to the correspondence information between the input personnel and the collection role preset in the configuration information and the working time information of the input personnel;

[0123] The second determination module is used to determine the application interface of the automatic collection device corresponding to the second data element according to the second data element in the collection data element set collected by the automatic collection device indicated in the configuration information.

[0124] In one or more embodiments, the generating unit 706 includes:

[0125] A first processing module is used to send the first data element and the collection requirement of the first data element at any information collection moment in the information collection task to the large language model, call the large language model to generate an information collection instruction for the information entry personnel, and push the information collection instruction to the information entry personnel;

[0126] The first calling module is used to send the second data element and the collection requirement of the second data element at any information collection moment to the large language model, and call the large model to generate a calling instruction for the automatic collection device application interface.

[0127] In one or more embodiments, the generating unit 706 or the acquiring unit 708 includes:

[0128] A first sending generation module, configured to send a first collection session request of the current round to the information collection role, and generate a first collection session corresponding to the information collection task of the current round based on the first collection session request and a preset large language model;

[0129] A first receiving module, configured to receive first feedback information of an information collection role from the first collection session based on the first collection session;

[0130] An extraction module, used to extract data parameters corresponding to the current round in the collection data element set from the first feedback information;

[0131] A sending extraction module is used to send a second collection session request to the information collection role when the extracted data parameters do not meet the information collection requirements of the current round, and obtain second feedback information fed back by the information collection role based on the second collection session request; extract the data parameters corresponding to the current round in the collection data element set from the second feedback information;

[0132] Repeat the above steps until the extracted data parameters meet the information collection requirements of the current round.

[0133] In one or more embodiments, a first receiving module, or an extraction module, includes:

[0134] A first receiving subunit, configured to receive first feedback information from an information input person or an automatic collection device from a first collection session;

[0135] A sending subunit, configured to send data processing information to the large language model; the data processing information includes the first feedback information and data element items corresponding to the current round and the collection requirements of the data elements;

[0136] The second calling subunit is used to call the large language model to extract the corresponding data parameters of the current round in the collected data element set based on the data processing information.

[0137] In one or more embodiments, the second calling subunit includes:

[0138] The acquisition submodule is used to obtain the historical session data corresponding to the collection session of the historical round, and the historical data parameters extracted in each historical round;

[0139] The extraction submodule is used to call the large language model to extract the corresponding data parameters of the current round in the collection data element set from the historical session data, historical data parameters and data processing information.

[0140] In one or more embodiments, the information collection device further includes:

[0141] A checking unit, used to check the quality of the feedback information according to the preset correlation logic between different data items and the background knowledge base;

[0142] The second determination unit is used to determine that the extracted data parameters do not meet the information collection requirements of the current round when there are illogical data items in the feedback information.

[0143] In one or more embodiments, the information collection device further includes:

[0144] A third determination unit is used to build a context environment based on the feedback information according to the preset correlation logic between different data items and the background knowledge base, and determine the data items that are missed or illogical in the feedback information; the feedback information includes at least one of the first feedback information and the second feedback information;

[0145] The reasoning supplement and correction unit is used to supplement or correct missed or illogical data items based on reasoning.

[0146] In one or more embodiments, the first determining unit 710 includes:

[0147] The encoding unit is used to encode the data parameters into information collection results in XML format or JSON format.

[0148] According to another aspect of the embodiment of the present application, an electronic device for implementing the above-mentioned information collection method is also provided. The electronic device may be Figure 8 The terminal device or server shown in the figure. This embodiment is described by taking the electronic device as a server as an example. Figure 8As shown, the electronic device includes a memory 802 and a processor 804. The memory 802 stores a computer program, and the processor 804 is configured to execute the steps in any of the above method embodiments through the computer program.

[0149] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0150] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0151] S1, obtain the information collection task and generate a collection data element set corresponding to the information collection task;

[0152] S2, determining the information collection role corresponding to each data element in the collection data element set, and configuring and generating a collection session request; the information collection role includes information entry personnel and / or information collection equipment;

[0153] S3, sending a collection session request to the information collection role, and generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model;

[0154] S4, obtaining data parameters corresponding to the collected data element set based on the collected session;

[0155] S5, encoding the data parameters into information collection results in a preset format, and determining that the information collection task is completed.

[0156] Alternatively, a person skilled in the art may understand that: Figure 8 The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (MID), a PAD, and other terminal devices. Figure 8 The electronic device and the electronic equipment described above are not limited in structure. Figure 8 More or fewer components (such as network interfaces, etc.) as shown in, or with Figure 8 Different configurations are shown.

[0157] Among them, the memory 802 can be used to store software programs and modules, such as the program instructions / modules corresponding to the information collection method and device in the embodiments of the present application. The processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 802, that is, to implement the above-mentioned information collection method. The memory 802 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 802 may further include a memory remotely located relative to the processor 804, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 802 can be specifically, but not limited to, used to store information collection results. As an example, such as Figure 8 As shown, the memory 802 may include, but is not limited to, the acquisition generation unit 702, the determination configuration unit 704, the generation unit 706, the acquisition unit 708, and the determination unit 710 in the information collection device. In addition, it may also include, but is not limited to, other module units in the information collection device, which will not be repeated in this example.

[0158] Optionally, the transmission device 806 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wired network and a wireless network. In one example, the transmission device 808 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 806 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0159] In addition, the electronic device mentioned above further includes: a display 808 for displaying the information collection result; and a connection bus 810 for connecting various module components in the electronic device mentioned above.

[0160] In other embodiments, the terminal device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. Among them, the nodes may form a peer-to-peer (P2P, Peer To Peer) network, and any form of computing device, such as a server, terminal and other electronic devices, may become a node in the blockchain system by joining the peer-to-peer network.

[0161] According to one aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, 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 performs the above-mentioned information collection method, wherein the computer program is configured to perform the steps of any one of the above-mentioned method embodiments when it is run.

[0162] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0163] S1, obtain the information collection task and generate a collection data element set corresponding to the information collection task;

[0164] S2, determining the information collection role corresponding to each data element in the collection data element set, and configuring and generating a collection session request; the information collection role includes information entry personnel and / or information collection equipment;

[0165] S3, sending a collection session request to the information collection role, and generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model;

[0166] S4, obtaining data parameters corresponding to the collected data element set based on the collected session;

[0167] S5, encoding the data parameters into information collection results in a preset format, and determining that the information collection task is completed.

[0168] Optionally, in this embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing hardware related to the terminal device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0169] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0170] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which can be personal computers, servers or network devices, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention.

[0171] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0172] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0173] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0175] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An information collection method, characterized in that: include: Acquire an information collection task, and generate a collection data element set corresponding to the information collection task; Determine the information collection role corresponding to each data element in the collection data element set, and configure and generate a collection session request; The information collection roles include information entry personnel and / or information collection equipment; Sending a collection session request to the information collection role, and generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model; Acquire data parameters corresponding to the collection data element set based on the collection session; The data parameters are encoded into an information collection result in a preset format, and the information collection task is determined to be completed.

2. The method according to claim 1, characterized in that The information acquisition task includes: Acquire a first type of information collection task, where the first type of information collection task is a collection task preset by the system, and the preset collection task has fixed information collection parameters; the information collection parameters include at least one of an identification, a life cycle, and a collection data element set of the information collection task; Alternatively, a second type of information collection task is obtained, where the second type of information collection task is a task generated by the large language model based on a collection task instruction described in text.

3. The method according to claim 1 or 2, characterized in that: The determining the information collection role corresponding to each data element in the collection data element set includes: Determine the input personnel corresponding to the first data element in the collection data element set according to the correspondence information between the input personnel and the collection role preset in the configuration information, and the working time information of the input personnel; According to the second data element in the set of collected data elements indicated in the configuration information to be collected by the automatic collection device, an application interface of the automatic collection device corresponding to the second data element is determined.

4. The method according to claim 3, characterized in that The generating the collection session corresponding to the information collection task based on the collection session request and the preset large language model includes: For any information collection moment in the information collection task, the first data element of any information collection moment and the collection requirement of the first data element are sent to the large language model, the large language model is called to generate an information collection instruction for the information entry personnel, and the information collection instruction is pushed to the information entry personnel; The second data element at any information collection moment and the collection requirement of the second data element are sent to the large language model, and the large model is called to generate a calling instruction for the application interface of the automatic collection device.

5. The method according to claim 1, characterized in that sending a collection session request to the information collection role, and generating a collection session corresponding to the information collection task based on the collection session request and a preset large language model; Acquiring data parameters corresponding to the collection data element set based on the collection session includes: Sending a first collection session request of the current round to the information collection role, and generating a first collection session corresponding to the information collection task of the current round based on the first collection session request and a preset large language model; Based on the first collection session, receiving first feedback information of the information collection role from the first collection session; Extracting data parameters corresponding to the current round in the collection data element set from the first feedback information; When the extracted data parameters do not meet the information collection task requirements of the current round, a second collection session request is sent to the information collection role, and second feedback information fed back by the information collection role is obtained based on the second collection session request; Extracting data parameters corresponding to the current round in the collection data element set from the second feedback information; Repeat the above steps until the extracted data parameters meet the information collection task requirements of the current round.

6. The method according to claim 5, characterized in that receiving first feedback information of the information collection role from the first collection session based on the first collection session; Extracting data parameters corresponding to the current round in the collection data element set from the first feedback information includes: receiving first feedback information from an information input person or an automatic collection device from the first collection session; Sending data processing information to the large language model; the data processing information includes the first feedback information and the data element items corresponding to the current round and the collection requirements of the data elements; Based on the data processing information, the large language model is called to extract data parameters corresponding to the current round in the set of collected data elements.

7. The method according to claim 6, characterized in that The step of calling the large language model to extract data parameters corresponding to the current round in the set of collected data elements based on the data processing information further includes: Obtaining historical session data corresponding to the collection session of the historical round, and historical data parameters extracted from each of the historical rounds; The large language model is called to extract data parameters corresponding to the current round in the collection data element set from historical session data, historical data parameters and the data processing information.

8. The method according to claim 5, characterized in that The method further comprises: Perform quality checks on feedback information based on the preset correlation logic between different data items and the background knowledge base; In the case where there are missed or illogical data items in the feedback information, it is determined that the extracted data parameters do not meet the information collection task of the current round.

9. The method according to claim 5, characterized in that The method further comprises: According to the preset correlation logic between different data items and the background knowledge base, a context environment is constructed based on the feedback information to determine the data items that are missed or illogical in the feedback information; the feedback information includes at least one of the first feedback information and the second feedback information; Based on reasoning, supplement or correct missed or illogical data items.

10. An information collection device, characterized in that: include: An acquisition generating unit, used to acquire an information collection task and generate a collection data element set corresponding to the information collection task; A determination configuration unit, used to determine the information collection role corresponding to each data element in the collection data element set, and configure and generate a collection session request; The information collection roles include information entry personnel and / or information collection equipment; A generating unit, configured to send a collection session request to the information collection role, and generate a collection session corresponding to the information collection task based on the collection session request and a preset large language model; An acquisition unit, configured to acquire data parameters corresponding to the collection data element set based on the collection session; The first determining unit is used to encode the data parameters into an information collection result in a preset format to determine that the information collection task is completed.

11. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 9 through the computer program.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 9 when executed.

Images