Intelligent inference-based multi-specialty artificial intelligence service collaborative decision-making and unified interaction method and platform

Abstract

This invention discloses a multi-disciplinary artificial intelligence service collaborative decision-making and unified interaction method and platform based on intelligent reasoning, relating to the field of artificial intelligence technology. It includes basic information acquisition: obtaining the problem to be processed to obtain the target problem. This invention uses a set ambiguity analysis method to determine whether the target problem has other meanings, so as to fully study and judge the target problem, and improve the accuracy of subsequent target problem output. A joint processing method is used to connect the target problem with the problems before it is proposed for contextual analysis, so as to accurately output the result of the target problem required by the user. A set reasoning generation method is used to generate a target path based on the domain described by keywords in the determined information or target information, realizing a transparent and interpretable reasoning path. Furthermore, a storage establishment method is used to store the target path. The overall process is adaptable to problems in different domains, and the stored target path can be used to adapt to different scenarios.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a method and platform for collaborative decision-making and unified interaction of multi-disciplinary artificial intelligence services based on intelligent reasoning. Background Technology

[0002] Multi-disciplinary AI service collaboration refers to the integration of specialized AI subsystems from different fields (such as healthcare, finance, engineering, and law) through a unified collaborative framework for task decomposition, information alignment, reasoning fusion, and interactive presentation. It enables collaborative reasoning, conflict detection, and evidence fusion in cross-domain tasks, and improves the accuracy, transparency, and interpretability of decision-making through unified user interaction, traceable reasoning chains, and rigorous data governance, while reducing the cognitive cost and operational complexity for users.

[0003] Patent publication number CN120653510A discloses an AI-based automated planning and integration service system. This AI-based automated planning and integration service system includes: a multi-source heterogeneous data access and processing module for real-time preprocessing of multi-source heterogeneous data from sensors, historical business logs, and third-party APIs; a unified data modeling and storage module for semantic modeling and storage of the multi-source heterogeneous data using a knowledge representation model; and an intelligent planning and scheduling module for reasoning and optimizing timeliness, cost, and resource utilization. This AI-based automated planning and integration service system can continuously refine decision-making strategies, ensuring that optimal or near-optimal scheduling solutions are always provided in complex and ever-changing business environments, thereby significantly improving overall operational efficiency and resource utilization.

[0004] In existing technologies, inputs and outputs from different domains cannot be accurately mapped, and the reasoning process is prone to ambiguity or inconsistency, making it difficult to achieve transparent and explainable reasoning paths. The unified processing mechanism for uncertain and conflicting information is inadequate, easily leading to conflicts and a lack of reliable resolution strategies. Real-time performance and scalability are limited, especially in high-concurrency scenarios and large-scale knowledge graphs, where costs are high, response is slow, and there is a lack of end-to-end adaptive capabilities from ambiguity identification to resource scheduling. This results in slow implementation speed and insufficient trust in new scenarios, requiring extensive customized development and continuous maintenance. Therefore, this invention is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a method and platform for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-professional artificial intelligence service collaborative decision-making and unified interaction method based on intelligent reasoning, the method comprising:

[0007] Basic information acquisition: Obtain the problem to be addressed to obtain the target problem; obtain the multi-disciplinary artificial intelligence and its service fields to be used to obtain artificial intelligence and field information;

[0008] Basic information processing: Establish the relationship between artificial intelligence and domain information, and integrate artificial intelligence, domain information and the relationship to obtain a reference catalog;

[0009] Information ambiguity analysis: Based on the target question, the ambiguity analysis method is used to determine whether the target question has other meanings, and the judgment result and the information to be processed are obtained. The information to be processed is the target information and the derived information or the target information. When the judgment result is that the target question has other meanings, the information to be processed is the target information and the derived information. When the judgment result is that the target question does not have other meanings, the information to be processed is the target information.

[0010] Analysis results determination: Based on the target information and derived information, the processing direction and determination information are determined by the information determination method according to the context of the target information. When the processing direction is fuzzy output, the determination information is the target information and derived information. When the processing direction is precise output, the determination information is the selected information.

[0011] Reasoning path determination: Based on known information or target information, the generation path of information results is determined through reasoning generation methods to obtain the target path;

[0012] Problem output: Based on the target path and reference directory, the problem results are obtained by processing the defined or target information using information processing methods.

[0013] Path storage: A path repository is established through a storage creation method. The path repository provides target paths to assist in subsequent determination of information or target information.

[0014] Furthermore, the ambiguity analysis method includes: extracting keywords based on the target question to obtain selected keywords; dividing the target question into paragraphs based on the selected keywords to obtain sub-question segments; establishing a correspondence between sub-question segments and selected keywords; determining the target domain based on the selected keywords; obtaining a judgment result based on the number of target domains in all sub-question segments; integrating the target domain, selected keywords, and sub-question segments based on the correspondence to obtain sub-target information; integrating the sub-target information to obtain target information; when the number of target domains in all sub-question segments is single, the judgment result indicates that the target question has no other meaning, and the target information is information to be processed; when the number of target domains in sub-question segments is multiple, the judgment result indicates that the target question has other meanings; recording the sub-question segments with multiple target domains to obtain replacement items; removing the target domains from the target information based on the replacement items to obtain the remaining domains; replacing the target domains in the replacement items with the remaining domains to obtain sub-derived information; replacing the replacement items in the target information based on the sub-derived information to obtain derived information; and the target information and derived information are information to be processed.

[0015] Furthermore, the process of dividing the target question into sub-question segments based on selected keywords is as follows: determine the number of selected keywords in the target question to obtain a quantity result. When the quantity result indicates that the number of selected keywords in the target question is a single keyword, the target question is a sub-question segment. When the quantity result indicates that the number of selected keywords in the target question is multiple, determine the continuity of the multiple selected keywords. When the multiple selected keywords are not continuous, divide the target question into sub-question segments directly based on the position of the selected keywords. When the multiple selected keywords are continuous, repeat the sentence structure in the target question for a single selected keyword to obtain a sub-question segment.

[0016] Furthermore, the joint processing method includes: presetting a processing time period; obtaining past target problems processed during the processing time period to obtain past problems; obtaining the domains of past problems to obtain related domains; determining whether there are past target problems processed during the processing time period to obtain processing direction and determination information; when there are past target problems processed during the processing time period, extracting information corresponding to the domains and related domains from the target information and derived information to obtain selection information; the processing direction is precise output, and the selection information is determination information; when there are no past target problems processed during the processing time period, the processing direction is fuzzy output, and the target information and derived information are determination information.

[0017] Furthermore, the reasoning generation method includes: determining that one of the information and the target information is the information to be reasoned; performing information matching based on the domain and domain information in the information to be reasoned to obtain a matching result; the matching result is the artificial intelligence corresponding to the domain of the information to be reasoned; and recording the matching results of all domain and domain information in the information to be reasoned to obtain the target path.

[0018] Furthermore, the information processing method includes: obtaining the artificial intelligence to be used based on the target path and the reference directory; when processing certain information, obtaining result information based on the certain information, the artificial intelligence to be used, and the target path; integrating the result information and the processing direction of the certain information to obtain the problem result; and when processing target information, obtaining the problem result based on the target information, the artificial intelligence to be used, and the target path.

[0019] Furthermore, the storage establishment method includes: establishing a path repository for storing target paths and corresponding determination information or target information; in the subsequent processing of determination information or target information, determining the target path and obtaining a reference path by the number of fields in the subsequent determination information or target information in the path repository; and replacing the fields in the reference path with the specific fields in the subsequent determination information or target information based on the reference directory to complete the auxiliary provision of the target path.

[0020] The multi-disciplinary AI service collaborative decision-making and unified interaction platform based on intelligent reasoning uses the aforementioned multi-disciplinary AI service collaborative decision-making and unified interaction method based on intelligent reasoning.

[0021] Compared with the prior art, the beneficial effects of the present invention are:

[0022] This multi-disciplinary AI service collaborative decision-making and unified interaction method and platform based on intelligent reasoning uses an ambiguity analysis method to determine whether the target question has other meanings, so as to fully study and judge the target question and improve the accuracy of the subsequent target question output. It uses a joint processing method to connect the target question with the question before it is raised for contextual analysis, so as to accurately output the target question required by the user. It uses a reasoning generation method to generate the target path based on the domain described by the keywords in the determined information or target information, so as to realize a transparent and interpretable reasoning path. Furthermore, it uses a storage establishment method to store the target path. The whole process is easy to adapt to problems in different domains, and the stored target path can be used to adapt to different scenarios.

[0023] Meanwhile, in the joint processing method, the setting of the processing time period ensures that the determination of the processing direction and the identification of information are timely, avoiding errors in the identification of information and processing direction due to lack of timeliness. Precise output indicates that there is target information and derived information in the same field as the previous problem during the processing time period, while fuzzy output indicates that there is no target information and derived information in the same field as the previous problem during the processing time period. By setting the processing direction, it is convenient to provide prompts when outputting information, so as to improve the richness and accuracy of information generation.

[0024] Simultaneously, by setting the storage establishment method, a path repository is established. The path repository is used to store the target path and the corresponding definite information or target information. In the subsequent target path, the corresponding sub-problem segments can be brought into the corresponding positions according to the domain location. In specific use, the corresponding multi-disciplinary artificial intelligence can be connected according to the domain location in the target path, that is, a template for solving the problem can be generated, which helps to improve the efficiency of multi-disciplinary artificial intelligence in handling problems. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0026] Figure 2 This is a schematic diagram of the processing time period structure of the present invention;

[0027] Figure 3 This is a schematic diagram of the segmented structure of the target problem in this invention;

[0028] Figure 4 This is a schematic diagram of the structure of the artificial intelligence processing sub-problem segment of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] A collaborative decision-making and unified interaction method for multi-disciplinary AI services based on intelligent reasoning integrates specialized AI subsystems from different fields within a unified platform through task decomposition, information alignment, reasoning fusion, and interpretable interactive presentation. It achieves conflict detection, weight allocation, and result fusion of multi-disciplinary outputs through cross-domain knowledge representation, collaborative reasoning frameworks, and evidence chain management. Furthermore, it enhances the accuracy, transparency, and usability of decision-making through unified human-computer interaction, traceable reasoning paths, and robust data governance, while reducing the cognitive burden and operational complexity for users.

[0031] like Figures 1-4 As shown, this invention provides a technical solution: a multi-professional artificial intelligence service collaborative decision-making and unified interaction method based on intelligent reasoning, the method comprising:

[0032] Basic information acquisition: Obtain the problem to be addressed to obtain the target problem; obtain the multi-disciplinary artificial intelligence and its service fields to be used to obtain artificial intelligence and field information;

[0033] It's important to note that the target problem is the issue that needs to be addressed. This is obtained by establishing a connection with users, specifically by receiving user queries (the target problem) and acquiring the necessary multi-disciplinary AI and its service areas. This process involves obtaining the AI ​​to be used and the different domains it addresses. For example, in the medical field, the target AI might be medical diagnostics / radiology / pathology AI. Typical inputs include patient symptom descriptions, signs, lab results, and images or pathology slides. Core outputs include a list of possible diagnoses, the confidence level of each diagnosis, key evidence fragments, differential diagnoses to be ruled out, and recommended next steps. In the financial risk control field, the target AI might be financial risk control models / fraud detection AI. Typical inputs include transaction records, account behavior logs, credit reports, and geographic and temporal features. Core outputs include risk scores and abnormal transaction indicators. For AI applications in engineering design, the following are targeted: Structural engineering optimization AI and manufacturing process optimization AI. Typical inputs include material parameters, load conditions, geometric constraints, and cost and schedule targets. Core outputs include optimized design schemes, performance predictions, cost / weight / reliability trade-offs, and recommended process parameters. In the natural language processing field, the following are targeted: Dialogue systems / knowledge graph question-answering AI. Typical inputs include user questions, knowledge base entries, common questions and answers, and contextual dialogue history. Core outputs include answer text, relevant evidence links, questions requiring clarification, and candidate next steps. In the energy and environment field, the following are targeted: Climate models / energy system optimization AI. Typical inputs include historical meteorological data, energy demand, equipment operation data, and policy constraints. Core outputs include future trend predictions, optimal operating strategies, risk assessments, and scenario analysis.

[0034] Basic information processing: Establish the relationship between artificial intelligence and domain information, and integrate artificial intelligence, domain information and the relationship to obtain a reference catalog;

[0035] It is important to note that the process of establishing the relationship between artificial intelligence and domain information involves labeling artificial intelligence with a tag that fits its work domain. This relationship facilitates the subsequent search for corresponding artificial intelligence for decision support through domain matching. By using a reference directory to include artificial intelligence, domain information, and relationships, it is easier to find information in a unified manner later.

[0036] Information ambiguity analysis: Based on the target question, the ambiguity analysis method is used to determine whether the target question has other meanings, and the judgment result and the information to be processed are obtained. The information to be processed is the target information and the derived information or the target information. When the judgment result is that the target question has other meanings, the information to be processed is the target information and the derived information. When the judgment result is that the target question does not have other meanings, the information to be processed is the target information.

[0037] It is important to note that by using the ambiguity analysis method, we can determine whether the target question has other meanings. If other meanings exist, we can derive additional information from the target question to fully study and judge the target question, thereby improving the accuracy when outputting the target question in the future.

[0038] Analysis results determination: Based on the target information and derived information, the processing direction and determination information are determined by the information determination method according to the context of the target information. When the processing direction is fuzzy output, the determination information is the target information and derived information. When the processing direction is precise output, the determination information is the selected information.

[0039] It is important to note that by using the set joint processing method to connect the target question with the question before the target question is raised, and to perform context analysis to determine the accurate target question among target questions with other meanings, it is convenient to accurately output the results of the target question required by the user. By adding processing directions, multiple interfaces are provided when it is impossible to determine whether it is accurate, namely accurate output and fuzzy output. Accurate output is the information determined after context analysis, while fuzzy output is the uncertain information. At the same time, the target information when the judgment result feedback is that the target question does not have other meanings can also be understood as accurate output.

[0040] Reasoning path determination: Based on known information or target information, the generation path of information results is determined through reasoning generation methods to obtain the target path;

[0041] It should be noted that the target path is generated by using the inference generation method set up based on the domain described by the keywords in the determined information or target information. The target path is related to the number of segments in the determined information or target information.

[0042] Problem output: Based on the target path and reference directory, the problem results are obtained by processing the defined or target information using information processing methods.

[0043] It is important to note that the information processing method is set up to select the corresponding artificial intelligence based on the target path and reference directory to process the determined information or target information to obtain the problem result. That is, the process of obtaining the result of the target problem is the process of importing the target problem into the artificial intelligence to obtain the result.

[0044] Path storage: A path repository is established through a storage creation method. The path repository provides target paths to assist in subsequent determination of information or target information.

[0045] It is important to note that by setting the storage method, a path repository storage path is established so that subsequent direct replacement of the artificial intelligence and information to be processed in the path can provide a target path to assist in determining the information or target information.

[0046] like Figure 1 As shown, the ambiguity analysis method includes: extracting keywords based on the target question to obtain selected keywords; dividing the target question into paragraphs based on the selected keywords to obtain sub-question segments; establishing a correspondence between sub-question segments and selected keywords; determining the target domain based on the selected keywords; obtaining a judgment result based on the number of target domains in all sub-question segments; integrating the target domain, selected keywords, and sub-question segments based on the correspondence to obtain sub-target information; integrating the sub-target information to obtain target information; when the number of target domains in all sub-question segments is single, the judgment result indicates that the target question has no other meaning, and the target information is information to be processed; when the number of target domains in sub-question segments is multiple, the judgment result indicates that the target question has other meanings; recording the sub-question segments with multiple target domains to obtain replacement items; removing the target domains from the target information based on the replacement items to obtain the remaining domains; replacing the target domains in the replacement items with the remaining domains to obtain sub-derived information; replacing the replacement items in the target information based on the sub-derived information to obtain derived information; and the target information and derived information are information to be processed.

[0047] It's important to note that the process of extracting keywords based on the target question involves extracting keywords from a passage describing the target question. This can be done automatically by training a keyword extraction model or manually. The process of determining the target domain based on the selected keywords involves identifying the domain described by the selected keywords. This can also be done automatically by training a model related to keyword attribution or manually. For example, if the target question is "I will describe artificial intelligence in a passage," the selected keyword is "artificial intelligence," and the target domain is the domain related to artificial intelligence. The process of determining the result based on the number of target domains for all sub-question segments involves segmenting all sub-question segments from the target question. If a single sub-question segment has multiple target domains, it indicates that the target question exists. Other meanings are possible, and conversely, if there are no other meanings, it means that the target problem does not have multiple meanings. The process of recording the replacement items for sub-problem segments with multiple target domains is as follows: that is, locating the sub-problem segments that need to be replaced and adding different domains to them. There may be multiple remaining domains. Specifically, each sub-problem segment contains only one domain. By replacing the target domain in the replacement item with the remaining domain, we obtain sub-derived information. The sub-derived information is then used to replace the replacement item in the target information to obtain derived information. The number of derived information corresponds to the number of target domains. Derived information can be understood as a derivative of the target information. The difference between it and the target information is that the target domains of some sub-problem segments are different, that is, the specialization of artificial intelligence processing used later is also different. This is to deal with the problem that the target problem may have multiple meanings and improve the richness of subsequent information output.

[0048] like Figure 1 and Figure 3 As shown, the process of dividing the target question into sub-question segments based on selected keywords is as follows: The number of selected keywords in the target question is determined to obtain a quantity result. When the quantity result indicates that the number of selected keywords in the target question is a single keyword, the target question is a sub-question segment. When the quantity result indicates that the number of selected keywords in the target question is multiple, the continuity of the multiple selected keywords is determined. When the multiple selected keywords are not continuous, the target question is directly divided into sub-question segments based on the position of the selected keywords. When the multiple selected keywords are continuous, the sentence structure in the target question is repeated for a single selected keyword to obtain a sub-question segment.

[0049] It's important to note that when the quantitative analysis results indicate a single keyword selected in the target question, the target question is considered a sub-question segment. This means the target question is a description of the selected keywords, and this is treated as a sub-question segment. When the quantitative analysis results indicate multiple keywords selected in the target question, the process of determining the continuity of these keywords involves checking the continuity of related terms. If they are not continuous, it means there are multiple paragraphs between the selected keywords, and paragraph segmentation is performed. If they are continuous, it means there is a connection between the selected keywords, i.e., the keywords are connected using the word "and". In this case, sub-question segments are obtained by repeating the sentence structure from the target question. For example, in the sentence "Introduce me to apples and computers," there are two keywords: "apple" and "computer." Sub-question segments are obtained by repeating the sentence structure from the target question, i.e., "Introduce me to apples" and "Introduce me to computers." This process of segmenting the target question into paragraphs based on the selected keywords allows different specialized AI systems to process the corresponding sub-question segments. Figure 3 The text indicates that a single target problem has 5 target domains, and the sub-problem segments correspond to the target domains, that is, there are 5 sub-problem segments.

[0050] like Figure 1 and Figure 2 As shown, the joint processing method includes: a preset processing time period; obtaining the previous target problem processed during the processing time period to obtain the previous problem; obtaining the domain of the previous problem to obtain the relevant domain; determining whether there is a previous target problem processed during the processing time period to obtain the processing direction and definite information; when there is a previous target problem processed during the processing time period, extracting the information corresponding to the domain and the relevant domain from the target information and the derived information to obtain the selection information; the processing direction is precise output, and the selection information is definite information; when there is no previous target problem processed during the processing time period, the processing direction is fuzzy output, and the target information and the derived information are definite information.

[0051] It's important to note that the processing time period should be set based on actual usage. Specifically, it's a time frame, such as counting back one day from the time the target question is received. This range constitutes the processing time period. The exact length of the processing time period can be determined based on actual usage. Setting the processing time period ensures timeliness in determining the processing direction and identifying information, avoiding errors in information identification and processing direction due to a lack of timeliness. Precise output indicates the presence of target information and derived information in the same domain as previous questions during the processing time period, while fuzzy output indicates the absence of target information and derived information in the same domain as previous questions during the processing time period. By setting the processing direction, prompts can be provided during information output, improving the richness and accuracy of information generation. Figure 2As shown, Figure 2 In this text, 'a' represents the length of the processing time period, and 'b' represents the position of the processing time period on the timeline.

[0052] like Figure 1 As shown, the reasoning generation method includes: determining that one of the information and the target information is the information to be reasoned; performing information matching based on the domain and domain information in the information to be reasoned to obtain a matching result; the matching result is the artificial intelligence corresponding to the domain of the information to be reasoned; and recording the matching results of all domain and domain information in the information to be reasoned to obtain the target path.

[0053] It is important to note that the process of determining whether the information or the target information is the information to be reasoned about involves inferring either the information to be determined or the target information as the information to be reasoned about based on the process. The information to be reasoned about contains a domain. By matching the domain of the information to be reasoned about with the domain information, a matching result can be obtained. By recording all the domain matching results, the target path can be obtained. Subsequently, by associating the domain with the sub-problem segments, the corresponding sub-problem segments can be provided for solving the target path.

[0054] like Figure 1 As shown, the information processing method includes: obtaining the artificial intelligence to be used based on the target path and the reference directory; when processing certain information, obtaining result information based on the certain information, the artificial intelligence to be used, and the target path; integrating the result information and the processing direction of the certain information to obtain the problem result; and when processing target information, obtaining the problem result based on the target information, the artificial intelligence to be used, and the target path.

[0055] It is important to note that the information processing method, which generates results for specific or target information based on the known target path and reference directory, plays the role of information matching and work distribution. By processing and integrating the distributed information through artificial intelligence, the problem results can be obtained. At the same time, the setting of the processing direction in the problem results can provide the degree of precision or fuzziness of the information output.

[0056] like Figure 1 and Figure 4 As shown, the storage establishment method includes: establishing a path repository to store the target path and the corresponding determination information or target information; in the subsequent processing of the determination information or target information, determining the target path and obtaining the reference path by the number of fields in the subsequent determination information or target information in the path repository; and replacing the fields in the reference path with the specific fields in the subsequent determination information or target information based on the reference directory to complete the auxiliary provision of the target path.

[0057] It's important to note that the storage method settings are used to create a path repository. This repository stores target paths and corresponding definitional or target information. Subsequently, the sub-problem segments in the obtained target path are inserted into their corresponding positions based on the domain's location. In practice, the location of the domain within the target path can be used to integrate corresponding multi-disciplinary AI, generating problem-solving templates and improving the efficiency of multi-disciplinary AI in handling problems. Figure 4 This means generating a template for solving the problem; specifically... Figure 4 The target path expressed in the text can be used to process problems by replacing specific sub-problem segments and specific artificial intelligence.

[0058] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. A collaborative decision-making and unified interaction method for multi-professional artificial intelligence services based on intelligent reasoning, the method comprising: Basic information acquisition: Obtain the problem to be addressed to obtain the target problem; obtain the multi-disciplinary artificial intelligence and its service fields to be used to obtain artificial intelligence and field information; The method is characterized in that it further includes: Basic information processing: Establish the relationship between artificial intelligence and domain information, and integrate artificial intelligence, domain information and the relationship to obtain a reference catalog; Information ambiguity analysis: Based on the target question, the ambiguity analysis method is used to determine whether the target question has other meanings, and the judgment result and the information to be processed are obtained. The information to be processed is the target information and the derived information or the target information. When the judgment result is that the target question has other meanings, the information to be processed is the target information and the derived information. When the judgment result is that the target question does not have other meanings, the information to be processed is the target information. Analysis results determination: Based on the target information and derived information, the processing direction and determination information are determined by the information determination method according to the context of the target information. When the processing direction is fuzzy output, the determination information is the target information and derived information. When the processing direction is precise output, the determination information is the selected information. Reasoning path determination: Based on known information or target information, the generation path of information results is determined through reasoning generation methods to obtain the target path; Problem output: Based on the target path and reference directory, the problem results are obtained by processing the defined or target information using information processing methods. Path storage: A path repository is established through a storage creation method. The path repository provides target paths to assist in subsequent determination of information or target information.

2. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The ambiguity analysis method includes: extracting keywords based on the target question to obtain selected keywords; dividing the target question into paragraphs based on the selected keywords to obtain sub-question segments; establishing a correspondence between sub-question segments and selected keywords; determining the target domain based on the selected keywords; obtaining a judgment result based on the number of target domains in all sub-question segments; integrating the target domain, selected keywords, and sub-question segments based on the correspondence to obtain sub-target information; integrating the sub-target information to obtain target information; when the number of target domains in all sub-question segments is single, the judgment result indicates that the target question has no other meaning, and the target information is information to be processed; when the number of target domains in sub-question segments is multiple, the judgment result indicates that the target question has other meanings; recording the sub-question segments with multiple target domains to obtain replacement items; removing the target domains from the target information based on the replacement items to obtain the remaining domains; replacing the target domains in the replacement items with the remaining domains to obtain sub-derived information; replacing the replacement items in the target information based on the sub-derived information to obtain derived information; and the target information and derived information are information to be processed.

3. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The process of dividing the target question into sub-question segments based on selected keywords is as follows: First, determine the number of selected keywords in the target question. If the result indicates that the number of selected keywords is a single keyword, the target question is a sub-question segment. If the result indicates that the number of selected keywords is multiple, determine the continuity of the selected keywords. If the selected keywords are not continuous, divide the target question into sub-question segments directly based on the position of the selected keywords. If the selected keywords are continuous, repeat the sentence structure from the target question for a single selected keyword to obtain a sub-question segment.

4. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The joint processing method includes: presetting a processing time period; obtaining the previous target problem processed during the processing time period to obtain the previous problem; obtaining the domain of the previous problem to obtain the relevant domain; determining whether there is a previous target problem processed during the processing time period to obtain the processing direction and determination information; when there is a previous target problem processed during the processing time period, extracting the information corresponding to the domain and the relevant domain from the target information and the derived information to obtain the selection information; the processing direction is a precise output, and the selection information is determination information; when there is no previous target problem processed during the processing time period, the processing direction is a fuzzy output, and the target information and the derived information are determination information.

5. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The reasoning generation method includes: determining that one of the information and the target information is the information to be reasoned; performing information matching based on the domain and domain information in the information to be reasoned to obtain a matching result; the matching result is the artificial intelligence corresponding to the domain of the information to be reasoned; and recording the matching results of all domain and domain information in the information to be reasoned to obtain the target path.

6. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The information processing method includes: obtaining the artificial intelligence to be used based on the target path and reference directory; when processing certain information, obtaining result information based on the certain information, the artificial intelligence to be used, and the target path; integrating the result information and the processing direction of the certain information to obtain the problem result; and when processing target information, obtaining the problem result based on the target information, the artificial intelligence to be used, and the target path.

7. The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning according to claim 1, characterized in that: The storage establishment method includes: establishing a path repository to store target paths and corresponding determination information or target information; in the subsequent processing of determination information or target information, determining the target path and obtaining a reference path by the number of fields in the subsequent determination information or target information in the path repository; and replacing the fields in the reference path with the specific fields in the subsequent determination information or target information based on the reference directory to complete the auxiliary provision of the target path.

8. A multi-disciplinary artificial intelligence service collaborative decision-making and unified interaction platform based on intelligent reasoning, characterized in that: The method for collaborative decision-making and unified interaction of multi-professional artificial intelligence services based on intelligent reasoning, as described in any one of claims 1-7, was used.

Images