AI robot dialogue control method and system based on big data search

Abstract

The invention is suitable for the technical field of chat robots, and provides an AI robot dialogue control method and system based on big data search, and the method comprises the following steps: receiving group chat robot control information which comprises autonomous dialogue frequency and empty time; the group chat message is processed and identified, and when a calling signal exists in the group chat message and the autonomous dialogue frequency or the empty time is reached, the group chat robot makes a dialogue; when the calling signal exists, collecting a group chat message corresponding to the calling signal, and performing a conversation according to the command type of the group chat message; when the autonomous dialogue frequency is reached, collecting N recent group chat messages, and performing interactive dialogue according to the N recent group chat messages; and when the empty time is up, collecting the latest group chat message, and inputting the group chat message into the AI big data model to obtain dialogue reply content. Therefore, the triggering conditions are diversified, communication reply can be carried out for various scenes, and the interactivity of group chat is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of chatbots, and in particular to an AI robot dialogue control method and system based on big data search. Background Art

[0002] With the rapid development of artificial intelligence technology, various chat interaction robots have emerged in an endless stream. For example, chat software supports group chat robots, which can set specific times to automatically send specific messages, such as weather forecasts, news summaries, group announcements, etc.; when users send specific questions and @ the robot, the robot will immediately give corresponding answers or guidance; when new members join the group chat, the robot can automatically send a welcome message to increase the friendly atmosphere of the group. However, the interactivity of existing group chat robots is still relatively weak, and the triggering conditions are very simple, which cannot interact more efficiently and deeply with group members. Therefore, it is necessary to provide an AI robot dialogue control method and system based on big data search to solve the above problems. Summary of the invention

[0003] In view of the shortcomings of the prior art, the purpose of the present invention is to provide an AI robot dialogue control method and system based on big data search to solve the problems existing in the above-mentioned background technology.

[0004] The present invention is implemented as follows: an AI robot dialogue control method based on big data search, the method comprising the following steps: receiving group chat robot control information, wherein the group chat robot control information includes autonomous conversation frequency and pause time; Process and identify group chat messages, and enable the group chat robot to have a conversation when there is a call signal in the group chat message, the autonomous conversation frequency or idle time is reached; When there is a call signal, the group chat message corresponding to the call signal is collected, and a conversation is conducted according to the command type of the group chat message, which is an inquiry query, statistical classification or text summary; When the autonomous conversation frequency is reached, the most recent N group chat messages are collected and interactive conversations are conducted based on the most recent N group chat messages; When the idle time is reached, the latest group chat messages are collected and input into the AI ​​big data model to obtain the conversation reply content.

[0005] As a further solution of the present invention: the step of processing and identifying the group chat message specifically includes: Determine whether a group chat robot summoning character appears in the group chat message, and if so, determine that a summoning signal exists; Count the number of messages in the group chat message after the most recent group chat robot conversation, delete duplicate messages when counting, and when the number of messages reaches the preset number, determine that the autonomous conversation frequency has been reached; Determine whether the latest group chat message is a question sentence. If yes, time the sending time of the latest group chat message to obtain a timing duration. When the timing duration reaches a preset duration, determine that the pause time has been reached.

[0006] As a further solution of the present invention: the step of collecting the group chat message corresponding to the call signal and conducting a conversation according to the command type of the group chat message specifically includes: Collecting a group chat message corresponding to the summon signal, and determining a command type corresponding to the group chat message; When it is an inquiry, the group chat message is input into the AI ​​big data model to obtain the conversation reply content; When it is statistical classification, message backtracking is performed to determine the chat topic and the topic replies to the chat topic, and statistical classification information is obtained based on the topic replies; When summarizing the text, message backtracking is performed to determine the chat topic and the topic replies to the chat topic, and the topic replies are integrated based on the AI ​​big data model to obtain text summary information.

[0007] As a further solution of the present invention: the step of determining the command type corresponding to the group chat message specifically includes: Preprocessing the group chat message, the preprocessing including word segmentation and removal of stop words; Match the preprocessed group chat messages with the keyword library. Each command type corresponds to a keyword library. The matching result is output, and the command type corresponding to the group chat message is determined according to the matching result.

[0008] As a further solution of the present invention: the step of conducting a conversation based on the most recent N group chat messages specifically includes: Summarize the most recent N group chat messages and determine the type of the summarized message, where the type of the summarized message is inquiry, topic discussion, or cannot be determined; When it is an inquiry query, determine the corresponding query question, and input the query question into the AI ​​big data model to obtain the dialogue response content; When it is a topic discussion, determine the topic phrase to be discussed, and input the topic phrase into the AI ​​big data model to obtain the dialogue reply content; When you are unsure, obtain social hot topics and conduct guided interactions based on them.

[0009] As a further solution of the present invention: the step of conducting guided interaction based on social hot topics specifically includes: Determine group chat interest tags based on all historical topic phrases, and filter social hot topics based on group chat interest tags; Retrieve the article titles of the filtered social hot topics, and input the article titles into the group chat dialog box.

[0010] Another object of the present invention is to provide an AI robot dialogue control system based on big data search, the system comprising: A control parameter determination module, used to receive group chat robot control information, wherein the group chat robot control information includes autonomous conversation frequency and pause time; The group chat robot startup module is used to process and identify group chat messages. When the group chat message contains a call signal, reaches the autonomous conversation frequency or idle time, the group chat robot will have a conversation. A call signal trigger module is used to collect the group chat message corresponding to the call signal when there is a call signal, and conduct a conversation according to the command type of the group chat message, where the command type is inquiry query, statistical classification or text summary; The conversation frequency trigger module is used to collect the latest N group chat messages when the autonomous conversation frequency is reached, and conduct interactive conversations based on the latest N group chat messages; The idle time trigger module is used to collect the latest group chat messages when the idle time is reached, and input the group chat messages into the AI ​​big data model to obtain the dialogue reply content.

[0011] As a further solution of the present invention: the group chat robot startup module includes: A calling signal determination unit, used to determine whether a group chat robot calling character appears in the group chat message, and when it appears, determine that a calling signal exists; A conversation frequency determination unit is used to count the number of messages in the group chat message after the most recent group chat robot conversation, delete duplicate messages when counting, and determine that the autonomous conversation frequency has been reached when the number of messages reaches a preset number; The pause time determination unit is used to determine whether the latest group chat message is a question sentence. If it is, the sending time of the latest group chat message is timed to obtain the timing duration. When the timing duration reaches a preset duration, it is determined that the pause time has been reached.

[0012] As a further solution of the present invention: the calling signal triggering module includes: A command type determination unit, used to collect a group chat message corresponding to the summon signal, and determine the command type corresponding to the group chat message; An inquiry and query reply unit, used for inputting the group chat message into the AI ​​big data model to obtain a dialogue reply content when it is an inquiry and query; A statistical classification reply unit, used for performing message backtracking, determining a chat topic and a topic reply to the chat topic, and obtaining statistical classification information based on statistics of the topic reply when statistical classification is performed; The text summary reply unit is used to perform message backtracking, determine the chat topic and the topic reply to the chat topic when summarizing the text, and integrate the topic reply based on the AI ​​big data model to obtain text summary information.

[0013] As a further solution of the present invention: the command type determination unit includes: A message preprocessing subunit, used to preprocess the group chat message, the preprocessing including word segmentation and stop word removal; The keyword matching subunit is used to match the pre-processed group chat message with the keyword library. Each command type corresponds to a keyword library. The command type determination subunit is used to output a matching result and determine the command type corresponding to the group chat message according to the matching result.

[0014] Compared with the prior art, the present invention has the following beneficial effects: In the present invention, when there is a call signal in the group chat message, the autonomous conversation frequency or the idle time is reached, the group chat robot will be triggered to conduct a conversation. The triggering conditions are diverse, and communication replies can be made for a variety of scenarios. And the conversation will be conducted according to the command type of the group chat message to achieve personalized conversation. This personalized conversation method can better meet the diverse needs of users and improve user experience; when the autonomous conversation frequency is reached, an interactive conversation will be conducted based on the latest N group chat messages. In this way, the group chat robot can frequently join the group chat to interact with group members, which is more efficient and in-depth. When the idle time is reached, the latest group chat message will be automatically collected, and the group chat message will be input into the AI ​​big data model to obtain the conversation reply content, so as to avoid the group chat from falling into long-term silence, enhance the interactivity of the group chat, and stimulate the enthusiasm of users to participate and improve the activity of the group chat. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The figure is a flow chart of an AI robot dialogue control method based on big data search.

[0016] Figure 2 The present invention is a flowchart for processing group chat messages in an AI robot dialogue control method based on big data search.

[0017] Figure 3 The present invention is a flowchart of conducting dialogue according to command types in the AI ​​robot dialogue control method based on big data search.

[0018] Figure 4A flowchart for determining command types in an AI robot dialogue control method based on big data search.

[0019] Figure 5 The present invention is a flowchart of conducting a conversation based on N messages in an AI robot conversation control method based on big data search.

[0020] Figure 6 The flowchart of guided interaction in the AI ​​robot dialogue control method based on big data search.

[0021] Figure 7 This is a structural diagram of the AI ​​robot dialogue control system based on big data search. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0024] like Figure 1 As shown, an embodiment of the present invention provides an AI robot dialogue control method based on big data search, and the method includes the following steps: S100, receiving group chat robot control information, where the group chat robot control information includes autonomous conversation frequency and pause time; S200, processing and identifying the group chat message, and when the group chat message contains a call signal, reaches the autonomous conversation frequency or idle time, enabling the group chat robot to have a conversation; S300, when there is a call signal, collect the group chat message corresponding to the call signal, and conduct a conversation according to the command type of the group chat message, where the command type is inquiry query, statistical classification or text summary; S400, when the autonomous conversation frequency is reached, collect the latest N group chat messages, and conduct an interactive conversation based on the latest N group chat messages; S500, when the idle time is reached, the latest group chat message is collected, and the group chat message is input into the AI ​​big data model to obtain the conversation reply content.

[0025] In the embodiment of the present invention, first, the group owner or administrator needs to set parameters for the group chat robot and input the group chat robot control information. The set parameter items include the autonomous conversation frequency and the idle time. Then, the group chat robot will process and identify the group chat message. When the group chat message contains a call signal, reaches the autonomous conversation frequency or the idle time, the group chat robot will be automatically triggered to enable the group chat robot to have a conversation. Specifically, when there is a call signal, the group chat message corresponding to the call signal is collected, and a conversation is conducted according to the command type of the group chat message, and the command type is inquiry query, statistical classification or text summary. In this way, a personalized conversation can be conducted, and this personalized conversation method can better meet the diverse needs of users and improve user experience; when the autonomous conversation frequency is reached, the latest N group chat messages are collected, and an interactive conversation is conducted according to the latest N group chat messages, where N is a fixed value set in advance, for example, N is 10, so that the group chat robot can frequently join the group chat to interact with group friends, which is more efficient and in-depth; when the idle time is reached, the latest group chat message will be automatically collected, and the group chat message will be input into the AI ​​big data model to obtain the conversation reply content, so as to avoid the group chat from falling into long-term silence. The AI ​​big data model can select ChatGPT, Wenxin Yiyan, Doubao, iFlytek Spark, etc. In this way, the trigger conditions of the group chat robot in the embodiment of the present invention are diverse, and it can communicate and reply for a variety of scenarios, which can not only enhance the interactivity of the group chat, but also stimulate the enthusiasm of users to participate and improve the activity of the group chat.

[0026] like Figure 2 As shown, as a preferred embodiment of the present invention, the step of processing and identifying the group chat message specifically includes: S201, determining whether a group chat robot summoning character appears in the group chat message, and if so, determining that a summoning signal exists; S202, counting the number of messages in the group chat message after the most recent group chat robot conversation, deleting duplicate messages during the counting, and when the number of messages reaches a preset number, determining that the autonomous conversation frequency has been reached; S203, determining whether the latest group chat message is a question sentence, and if so, timing the sending time of the latest group chat message to obtain a timing duration, and when the timing duration reaches a preset duration, determining that the pause time has been reached.

[0027] In the embodiment of the present invention, it is necessary to determine the triggering conditions of the group chat robot. First, it is determined whether the group chat robot summoning character appears in the group chat message. For example, the summoning character is @group chat robot or the name of the group chat robot. When the summoning character appears, it is determined that there is a summoning signal. Then, the number of messages after the most recent group chat robot conversation in the group chat message is counted. When counting, it is necessary to delete duplicate messages. For example, if the content of five messages is "received", then the five messages are counted as one. When the number of messages reaches the preset number, it is assumed that the autonomous conversation frequency is reached. In addition, it is also determined whether the latest group chat message is a question. If it is, the sending time of the latest group chat message is then timed to obtain the timing duration. When the timing duration reaches the preset duration, it is determined that the idle time has been reached, indicating that at this time, a group friend has raised a question but no one has answered it. The group chat robot will intervene to answer, so that the user experience is better.

[0028] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of collecting the group chat message corresponding to the call signal and conducting a conversation according to the command type of the group chat message specifically includes: S301, collecting a group chat message corresponding to the summon signal, and determining a command type corresponding to the group chat message; S302, when it is an inquiry, inputting the group chat message into the AI ​​big data model to obtain the dialogue reply content; S303, when it is statistical classification, message backtracking is performed to determine the chat topic and the topic replies to the chat topic, and statistical classification information is obtained based on the topic replies; S304, when it is a text summary, message backtracking is performed to determine the chat topic and the topic reply to the chat topic, and the topic reply is integrated based on the AI ​​big data model to obtain text summary information.

[0029] In the embodiments of the present invention, the group chat message corresponding to the summons signal is determined, and the command type corresponding to the group chat message is automatically determined. When the command type is an inquiry query, inputting the corresponding group chat message into the AI big data model will directly obtain the dialogue reply content. When the command type is statistical classification, message backtracking is performed to determine the chat topic and the topic reply to the chat topic. For example, the chat topic is: "Vote on the team building plan, deduct 1 for Plan 1 and deduct 2 for Plan 2", and the topic replies are: "1", "2", "1", "1", "2", "1", "1", "2". Then, statistical classification information is obtained based on the topic replies. For example, the statistical classification information is that five people choose Plan 1 and three people choose Plan 2, which is more rapid and convenient than the traditional group vote. When the command type is text summary, message backtracking is also performed to determine the chat topic and the topic reply to the chat topic. For example, the chat topic is: "Express opinions on the energy conservation and emission reduction project", and the topic reply is the reply content of the group members to this topic. Finally, text summary information is obtained by integrating the topic replies based on the AI big data model, and the reply content of each person is input into the AI big data model for optimization and summarization.

[0030] As Figure 4 shown, as a preferred embodiment of the present invention, the steps of determining the command type corresponding to the group chat message specifically include: S3011, preprocess the group chat message, and the preprocessing includes word segmentation and removal of stop words; S3012, match the preprocessed group chat message with the keyword library, and each command type corresponds to a keyword library; S3013, output the matching result, and determine the command type corresponding to the group chat message according to the matching result.

[0031] In the embodiments of the present invention, in order to better determine the command type, first, the group chat message is preprocessed. The natural language processing (NLP) library (such as the jieba tokenizer) is used to segment the message text and remove common stop words (such as "de", "le", "zai", etc.) to reduce noise. Then, the preprocessed group chat message is matched with the keyword library. Each command type corresponds to a keyword library, and the keyword library needs to be constructed in advance. For example, the keyword library for inquiry queries includes "query", "?", "why", "ma", etc., the keyword library for statistical classification includes "statistics", "classification", "proportion", etc., and the keyword library for text summary includes "summary", "generalization", "summarization", etc. The group chat message is character-matched with each keyword, and the matching result is output. According to the matching result, the command type corresponding to the group chat message can be automatically determined.

[0032] As Figure 5As shown, as a preferred embodiment of the present invention, the step of conducting a conversation based on the most recent N group chat messages specifically includes: S401, summarize the most recent N group chat messages and determine the type of the summarized message, where the type of the summarized message is inquiry, topic discussion, or cannot be determined; S402, when it is an inquiry query, determining a corresponding query question, and inputting the query question into the AI ​​big data model to obtain a dialogue response content; S403, when it is a topic discussion, determining the topic phrase to be discussed, and inputting the topic phrase into the AI ​​big data model to obtain the dialogue reply content; S404, when it cannot be determined, obtaining social hot topics, and conducting guided interaction based on the social hot topics.

[0033] In an embodiment of the present invention, the most recent N group chat messages are summarized to determine the type of the summarized message, which is an inquiry query, a topic discussion, or cannot be determined. Specifically, when a group chat message contains a question sentence, it is an inquiry query; when a certain amount of messages in the group chat message are discussing a certain topic, it is a topic discussion; otherwise, it cannot be determined. When it is an inquiry query, determine the corresponding query question, and input the query question into the AI ​​big data model to obtain the dialogue reply content. When it is a topic discussion, determine the topic phrase discussed, and input the topic phrase into the AI ​​big data model to obtain the dialogue reply content. When it cannot be determined, obtain real-time social hot topics, and conduct guided interaction based on social hot topics.

[0034] like Figure 6 As shown, as a preferred embodiment of the present invention, the step of conducting guided interaction based on social hot topics specifically includes: S4041, determining group chat interest tags based on all historical topic phrases, and screening social hot topics according to the group chat interest tags; S4042, retrieving the article titles of the filtered social hot topics, and inputting the article titles into the group chat dialog box.

[0035] In an embodiment of the present invention, the interest tags of the group chat are determined based on all historical topic phrases, for example, the interest tags are football, games, and cars, and then the social hot topics are filtered according to the group chat interest tags. The filtered social hot topics meet the tag content, and finally the relevant article titles of the filtered social hot topics are retrieved, and the article titles are entered into the group chat dialog box to trigger discussion among group members.

[0036] like Figure 7 As shown, an embodiment of the present invention further provides an AI robot dialogue control system based on big data search, the system comprising: A control parameter determination module 100, configured to receive group chat robot control information, wherein the group chat robot control information includes autonomous conversation frequency and pause time; The group chat robot activation module 200 is used to process and identify the group chat message, and when the group chat message contains a call signal, reaches the autonomous conversation frequency or idle time, the group chat robot conducts a conversation; The calling signal triggering module 300 is used to collect the group chat message corresponding to the calling signal when there is a calling signal, and conduct a conversation according to the command type of the group chat message, where the command type is inquiry query, statistical classification or text summary; The conversation frequency trigger module 400 is used to collect the latest N group chat messages when the autonomous conversation frequency is reached, and to conduct an interactive conversation based on the latest N group chat messages; The idle time trigger module 500 is used to collect the latest group chat messages when the idle time is reached, and input the group chat messages into the AI ​​big data model to obtain the dialogue reply content.

[0037] As a preferred embodiment of the present invention, the group chat robot starting module 200 includes: A calling signal determination unit, used to determine whether a group chat robot calling character appears in the group chat message, and when it appears, determine that a calling signal exists; A conversation frequency determination unit is used to count the number of messages in the group chat message after the most recent group chat robot conversation, delete duplicate messages when counting, and determine that the autonomous conversation frequency has been reached when the number of messages reaches a preset number; The pause time determination unit is used to determine whether the latest group chat message is a question sentence. If it is, the sending time of the latest group chat message is timed to obtain the timing duration. When the timing duration reaches a preset duration, it is determined that the pause time has been reached.

[0038] As a preferred embodiment of the present invention, the calling signal triggering module 300 includes: A command type determination unit, used to collect a group chat message corresponding to the summon signal, and determine the command type corresponding to the group chat message; An inquiry and query reply unit, used for inputting the group chat message into the AI ​​big data model to obtain a dialogue reply content when it is an inquiry and query; A statistical classification reply unit, used for performing message backtracking, determining a chat topic and a topic reply to the chat topic, and obtaining statistical classification information based on statistics of the topic reply when statistical classification is performed; The text summary reply unit is used to perform message backtracking, determine the chat topic and the topic reply to the chat topic when summarizing the text, and integrate the topic reply based on the AI ​​big data model to obtain text summary information.

[0039] As a preferred embodiment of the present invention, the command type determination unit includes: A message preprocessing subunit, used to preprocess the group chat message, the preprocessing including word segmentation and stop word removal; The keyword matching subunit is used to match the pre-processed group chat message with the keyword library. Each command type corresponds to a keyword library. The command type determination subunit is used to output a matching result and determine the command type corresponding to the group chat message according to the matching result.

[0040] The above only describes in detail the preferred embodiments of the present invention, which is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0041] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0042] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0043] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. An AI robot dialogue control method based on big data search, characterized in that: The method comprises the following steps: Receiving group chat robot control information, wherein the group chat robot control information includes autonomous conversation frequency and pause time; Process and identify group chat messages, and enable the group chat robot to have a conversation when there is a call signal in the group chat message, or when the autonomous conversation frequency or idle time is reached; When there is a call signal, the group chat message corresponding to the call signal is collected, and a conversation is conducted according to the command type of the group chat message, which is an inquiry query, statistical classification or text summary; When the autonomous conversation frequency is reached, the most recent N group chat messages are collected and interactive conversations are conducted based on the most recent N group chat messages; When the idle time is reached, the latest group chat messages are collected and input into the AI ​​big data model to obtain the conversation reply content.

2. The AI ​​robot dialogue control method based on big data search according to claim 1 is characterized in that: The step of processing and identifying the group chat message specifically includes: Determine whether a group chat robot summoning character appears in the group chat message, and if so, determine that a summoning signal exists; Count the number of messages in the group chat message after the most recent group chat robot conversation, delete duplicate messages when counting, and when the number of messages reaches the preset number, determine that the autonomous conversation frequency has been reached; Determine whether the latest group chat message is a question sentence. If yes, time the sending time of the latest group chat message to obtain a timing duration. When the timing duration reaches a preset duration, determine that the pause time has been reached.

3. The AI ​​robot dialogue control method based on big data search according to claim 1 is characterized in that: The step of collecting the group chat message corresponding to the call signal and conducting a conversation according to the command type of the group chat message specifically includes: Collecting a group chat message corresponding to the summon signal, and determining a command type corresponding to the group chat message; When it is an inquiry, the group chat message is input into the AI ​​big data model to obtain the conversation reply content; When it is statistical classification, message backtracking is performed to determine the chat topic and the topic replies to the chat topic, and statistical classification information is obtained based on the topic replies; When summarizing the text, message backtracking is performed to determine the chat topic and the topic replies to the chat topic, and the topic replies are integrated based on the AI ​​big data model to obtain text summary information.

4. The AI ​​robot dialogue control method based on big data search according to claim 3 is characterized in that: The step of determining the command type corresponding to the group chat message specifically includes: Preprocessing the group chat message, the preprocessing including word segmentation and removal of stop words; Match the pre-processed group chat messages with the keyword library. Each command type corresponds to a keyword library. The matching result is output, and the command type corresponding to the group chat message is determined according to the matching result.

5. The AI ​​robot dialogue control method based on big data search according to claim 1 is characterized in that: The step of conducting a conversation based on the most recent N group chat messages specifically includes: Summarize the most recent N group chat messages and determine the type of the summarized message, where the type of the summarized message is inquiry, topic discussion, or cannot be determined; When it is an inquiry query, determine the corresponding query question, and input the query question into the AI ​​big data model to obtain the dialogue response content; When it is a topic discussion, determine the topic phrase to be discussed, and input the topic phrase into the AI ​​big data model to obtain the dialogue reply content; When you are unsure, obtain social hot topics and conduct guided interactions based on them.

6. The AI ​​robot dialogue control method based on big data search according to claim 5 is characterized in that: The step of conducting guided interaction based on social hot topics specifically includes: Determine group chat interest tags based on all historical topic phrases, and filter social hot topics based on group chat interest tags; Retrieve the article titles of the filtered social hot topics, and input the article titles into the group chat dialog box.

7. AI robot dialogue control system based on big data search, characterized in that: The system comprises: A control parameter determination module, used to receive group chat robot control information, wherein the group chat robot control information includes autonomous conversation frequency and pause time; The group chat robot startup module is used to process and identify group chat messages. When the group chat message contains a call signal, reaches the autonomous conversation frequency or idle time, the group chat robot will have a conversation. A call signal trigger module is used to collect the group chat message corresponding to the call signal when there is a call signal, and conduct a conversation according to the command type of the group chat message, where the command type is inquiry query, statistical classification or text summary; The conversation frequency trigger module is used to collect the latest N group chat messages when the autonomous conversation frequency is reached, and conduct interactive conversations based on the latest N group chat messages; The idle time trigger module is used to collect the latest group chat messages when the idle time is reached, and input the group chat messages into the AI ​​big data model to obtain the dialogue reply content.

8. The AI ​​robot dialogue control system based on big data search according to claim 7 is characterized in that: The group chat robot startup module includes: A calling signal determination unit, used to determine whether a group chat robot calling character appears in the group chat message, and when it appears, determine that a calling signal exists; A conversation frequency determination unit is used to count the number of messages in the group chat message after the most recent group chat robot conversation, delete duplicate messages when counting, and determine that the autonomous conversation frequency has been reached when the number of messages reaches a preset number; The pause time determination unit is used to determine whether the latest group chat message is a question sentence. If it is, the sending time of the latest group chat message is timed to obtain the timing duration. When the timing duration reaches a preset duration, it is determined that the pause time has been reached.

9. The AI ​​robot dialogue control system based on big data search according to claim 7 is characterized in that: The calling signal triggering module comprises: A command type determination unit, used to collect a group chat message corresponding to the summon signal, and determine the command type corresponding to the group chat message; An inquiry and query reply unit, used for inputting the group chat message into the AI ​​big data model to obtain a dialogue reply content when it is an inquiry and query; A statistical classification reply unit, used for performing message backtracking, determining a chat topic and a topic reply to the chat topic, and obtaining statistical classification information based on statistics of the topic reply when statistical classification is performed; The text summary reply unit is used to perform message backtracking, determine the chat topic and the topic reply to the chat topic when summarizing the text, and integrate the topic reply based on the AI ​​big data model to obtain text summary information.

10. The AI ​​robot dialogue control system based on big data search according to claim 9 is characterized in that: The command type determination unit comprises: A message preprocessing subunit, used to preprocess the group chat message, the preprocessing including word segmentation and stop word removal; The keyword matching subunit is used to match the pre-processed group chat message with the keyword library. Each command type corresponds to a keyword library. The command type determination subunit is used to output a matching result and determine the command type corresponding to the group chat message according to the matching result.

Images