Artificial intelligence system for multi-domain questions and answers

Abstract

The invention discloses an artificial intelligence system for multi-domain questions and answers, which belongs to the technical field of multi-domain questions and answers and comprises an update analysis module, a supplementary acquisition module and a trust analysis module. The update analysis module is used for performing real-time update analysis on the knowledge data of the corresponding professional field and determining supplementary data of the corresponding professional field; the supplementary acquisition module is used for performing real-time data acquisition according to the supplementary data to obtain supplementary acquisition data of the corresponding professional field; the trust analysis module is used for performing real-time trust analysis on the supplementary collection data to obtain trust levels corresponding to the corresponding supplementary collection data, and the trust levels comprise an untrusted level, a trust qualification level and an accurate trust level; removing the supplementary collection data of which the trust level is an untrusted level, identifying potential knowledge data corresponding to the remaining supplementary collection data, marking the potential knowledge data as supplementary knowledge data, and marking a corresponding trust level label for the supplementary knowledge data; and storing the supplementary knowledge data in a knowledge base.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-domain question answering, and specifically relates to an artificial intelligence system for multi-domain question answering. Background Art

[0002] The multi-domain question answering system is an important branch in the field of artificial intelligence. Its core goal is to provide users with accurate and real-time cross-domain question answers by integrating multi-source heterogeneous knowledge. With the explosive growth of knowledge, such as a 10% annual increase in academic papers and millions of daily news event updates, the lag in domain knowledge update has become a key bottleneck restricting system performance. For example, if the side effects of new drugs in the medical field and breakthrough research results in the science and technology field are not incorporated into the knowledge base in a timely manner, it will lead to outdated or incorrect answers from the system, seriously affecting user trust.

[0003] Based on this, to solve the problem of multi-domain knowledge update, the present invention provides an artificial intelligence system for multi-domain question answering. Summary of the Invention

[0004] To solve the problems existing in the above solutions, the present invention provides an artificial intelligence system for multi-domain question answering.

[0005] The object of the present invention can be achieved by the following technical solutions: An artificial intelligence system for multi-domain question answering, comprising a knowledge base, an update analysis module, a supplementary acquisition module, a trust analysis module, and a question answering module; The knowledge base is used to store knowledge data in each professional field.

[0006] The update analysis module is used to perform real-time update analysis on the knowledge data in the corresponding professional field to determine the supplementary data in the corresponding professional field.

[0007] Further, performing real-time update analysis on the knowledge data in the corresponding professional field includes: Presetting the timeliness standards for each professional field, where the timeliness standards include a qualified standard and a qualification rate standard; Performing real-time analysis on the corresponding professional field according to the timeliness standards to obtain a supplementary analysis field; performing real-time analysis on the supplementary analysis field according to the timeliness standards to obtain the corresponding supplementary data.

[0008] Further, performing real-time analysis on the corresponding professional field according to the timeliness standards includes: Setting a retrieval unit, where the retrieval unit is used to retrieve in real time the knowledge association data not stored in the knowledge base in the corresponding professional field, and the knowledge association data includes retrieved knowledge data and verification result data of the corresponding retrieved knowledge data; Screening the knowledge association data according to the verification result data to obtain the field material data of the professional field; Obtain the knowledge update records in the corresponding professional fields in the knowledge base in real time; calibrate and analyze the material data in each field according to the knowledge update records and timeliness standards, and obtain the judgment results on whether the corresponding professional fields meet the timeliness standards.

[0009] Furthermore, calibrating and analyzing the material data in each field according to the knowledge update records and timeliness standards includes: Identify the update waiting duration corresponding to the material data in each field in real time according to the knowledge update records; summarize the update waiting durations of each professional field to form the update duration set of the professional field. Establish a field evaluation model, and the expression of the field evaluation model is: ; In the formula: (U, ST) is the input data, U is the update duration set of the corresponding professional field; ST is the timeliness standard of the corresponding professional field; U→ST indicates that the corresponding update duration set meets the requirements of the timeliness standard; the output data is the field evaluation value LP(U, ST), and the field evaluation value is 1 or 0. Perform real-time analysis on the update duration set and timeliness standard of the professional field according to the field evaluation model to obtain the corresponding field evaluation value. Mark the professional fields with a field evaluation value of 0 as supplementary analysis fields.

[0010] Furthermore, perform real-time analysis on the supplementary analysis fields according to the timeliness standards, including: Perform real-time retrieval on the supplementary analysis fields through the retrieval unit to obtain the knowledge retrieval data of the supplementary analysis fields; set the warning duration according to the timeliness standards. Obtain the verification result data and update waiting duration of the knowledge retrieval data in real time, and perform real-time screening on the knowledge retrieval data according to the verification result data. Perform real-time analysis on the remaining knowledge retrieval data according to the warning duration and update waiting duration. When the update waiting duration reaches the warning duration, supplement the knowledge retrieval data into the supplementary data of the supplementary analysis fields.

[0011] The supplementary acquisition module is used to perform real-time data acquisition according to the supplementary data to obtain the supplementary acquisition data of the corresponding professional field, and the supplementary acquisition data includes potential knowledge data and verification association data.

[0012] The trust analysis module is used to perform real-time trust analysis on the supplementary acquisition data to obtain the trust level corresponding to the corresponding supplementary acquisition data, and the trust levels include untrusted level, trust qualified level, and accurate trust level. Eliminate the supplementary acquisition data with an untrusted trust level, identify the potential knowledge data corresponding to the remaining supplementary acquisition data, mark the potential knowledge data as supplementary knowledge data, and mark the corresponding trust level label for the supplementary knowledge data; store the supplementary knowledge data in the knowledge base.

[0013] Furthermore, classify and store the supplementary knowledge data stored in the knowledge base according to the trust level, and the response priority of the accurate trust level is higher than that of the qualified trust level.

[0014] Furthermore, perform real-time trust analysis on the supplementary acquisition data, including: Preset the trust value intervals corresponding to each trust level, and the value range of the trust value is [0, 100]; Establish a trust value evaluation model, which is used to evaluate the trust value of the corresponding potential knowledge data; Real-time identify the verification associated data of the supplementary acquisition data, analyze the verification associated data through the trust value evaluation model, and obtain the trust value of the potential knowledge data corresponding to the supplementary acquisition data; Determine the trust level of the potential knowledge data according to the trust value and the trust value interval of the corresponding trust level.

[0015] Furthermore, update and adjust the trust level of the supplementary knowledge data stored in the knowledge base.

[0016] The Q&A module is used for users to ask questions and answers, real-time obtain the question data of the users, generate the corresponding response data according to the question data, and display the response data to the users.

[0017] Compared with the prior art, the beneficial effects of the present invention are: Through the mutual cooperation among the knowledge base, the update analysis module, the supplementary acquisition module, the trust analysis module and the Q&A module, the intelligent update of knowledge data in multiple professional fields is realized, the accuracy and timeliness of the knowledge data are guaranteed, and then the intelligent Q&A in multiple fields is realized; it can effectively cope with the challenges brought by the explosive growth of knowledge, ensure that key knowledge such as new drug side effects in the medical field and breakthrough research results in the scientific and technological field are updated to the knowledge base in time, avoid outdated or wrong answers caused by knowledge lag, and significantly improve the accuracy and reliability of cross-field Q&A; the output of accurate and real-time answers can directly enhance the user's trust in the multi-field Q&A system; users do not need to worry about decision-making biases caused by old or wrong knowledge. Especially in fields such as medicine and technology where the timeliness requirements of information are extremely high, the present invention can provide users with more valuable question answers, thereby optimizing the user interaction experience and enhancing user stickiness. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is the principle block diagram of the present invention. Specific embodiments

[0020] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0021] As Figure 1 shown, an artificial intelligence system for multi-domain question answering includes a knowledge base, an update analysis module, a trust analysis module, and a question answering module; The knowledge base is used to store knowledge data in each professional field.

[0022] The update analysis module is used to perform real-time update analysis on the knowledge data in the corresponding professional field, determine the supplementary data in the corresponding professional field, and the supplementary data is used to indicate the knowledge data that needs to be supplemented in this professional field.

[0023] In one embodiment, performing real-time update analysis on the knowledge data in the corresponding professional field includes: Identifying each professional field served by the question answering system, presetting the timeliness standards for each professional field. For example, when new knowledge data appears in a certain professional field, the requirement of its timeliness standard is to be stored in the knowledge base within one week, and a compliance rate of 90% is regarded as meeting the timeliness standard. That is, the timeliness standard includes two standards, one is the qualified standard, and the other is the qualification rate standard; because different professional fields have different requirements for the timeliness of knowledge answering, it is necessary to set the timeliness standard according to actual needs and make dynamic adjustments; specifically, it is set by the platform side.

[0024] Performing real-time analysis on the corresponding professional field according to the timeliness standard, judging whether the knowledge data update method in the corresponding professional field meets the requirements, and marking the professional field that does not meet the requirements as the supplementary analysis field; Performing real-time analysis on the supplementary analysis field according to the timeliness standard to obtain the corresponding supplementary data.

[0025] In one embodiment, performing real-time analysis on the corresponding professional field according to the timeliness standard includes: A retrieval unit is set up. The retrieval unit is used to retrieve the latest knowledge association data in the corresponding professional field in real time. The knowledge association data includes retrieved knowledge data and verification result data of the retrieved knowledge data. The verification result data proves whether it is knowledge data over time, that is, whether the knowledge data is true. Since there is a large amount of false data flooding current networks and other channels, in order to avoid affecting the accuracy of question answering due to rumor-mongering data, that is, the knowledge association data is determined in combination with subsequent actual verification results and has time persistence. It can be judged as long as the verification results determined from other subsequent sources are available. It is equivalent to retrieving various possible new knowledge data, marking them as retrieved knowledge data, and based on this retrieved knowledge data, obtaining the corresponding verification result data in real time. Integrating the retrieved knowledge data with verification result data into knowledge association data. Retrieving in real time in the above manner to obtain the knowledge association data in the corresponding professional field.

[0026] Screen the knowledge association data, eliminate the knowledge association data corresponding to the verification result data that fails the verification, and mark the remaining knowledge association data as domain material data.

[0027] Obtain the update record of the knowledge data in the corresponding professional field in the knowledge base in real time, and mark it as the knowledge update record. Perform calibration analysis on each domain material data according to the knowledge update record and the timeliness standard to obtain the judgment result on whether the corresponding professional field meets the timeliness standard.

[0028] In one embodiment, performing calibration analysis on each domain material data according to the knowledge update record and the timeliness standard includes: Determine the update waiting duration corresponding to each domain material data according to the knowledge update record, that is, the interval duration from the retrieved knowledge data of the determined domain material data to its update in the knowledge base. If there is no update or the obvious update waiting duration exceeds the timeliness standard, it is not necessary to wait for it to appear in the knowledge update record, and a non-compliant update waiting duration can be directly output. Generally, it is output with the duration just exceeding the timeliness standard. Summarize each update waiting duration corresponding to each professional field to form an update duration set. Establish a domain evaluation model. The domain evaluation model is used to analyze the update duration set of the corresponding professional field to judge whether it meets the timeliness standard. Specifically, calculate the qualification rate corresponding to the update duration set according to the qualified standard of the timeliness standard, and then judge according to the qualification rate standard of the timeliness standard. The expression of the domain evaluation model is: ; Where: (U, ST) is the input data, U is the set of update durations in the corresponding professional field; ST is the timeliness standard in the corresponding professional field; U→ST indicates that the corresponding set of update durations meets the requirements of the timeliness standard; the output data is the field evaluation value LP(U, ST), and the field evaluation value is 1 or 0; Perform real-time analysis on the set of update durations and the timeliness standard in the corresponding professional field according to the field evaluation model to obtain the corresponding field evaluation value; Mark the professional fields with a field evaluation value of 0 as supplementary analysis fields.

[0029] In one embodiment, perform calibration analysis on the material data of each field according to the knowledge update record and the timeliness standard, and can also perform analysis based on current technologies such as artificial intelligence. For example, establish a field calibration model based on a deep learning algorithm, and establish a corresponding training set through artificial means for training. The training set includes input data and output data. The input data is the knowledge update record, the field material data, and the timeliness standard, and the output data is the verification result, which is used to indicate whether the professional field is a supplementary analysis field; perform real-time analysis according to the field calibration model after successful training.

[0030] In one embodiment, perform real-time analysis on the supplementary analysis field according to the timeliness standard, including: Perform real-time retrieval on the supplementary analysis field through the retrieval unit in the above embodiment to obtain the corresponding knowledge retrieval data; set an early warning duration according to the timeliness standard, that is, how much the update waiting duration of the knowledge retrieval data reaches for early warning, and the early warning duration is less than or equal to the duration requirement corresponding to the timeliness standard; specifically set the early warning duration according to the requirements of the platform party. Generally, it is set to be less than the duration corresponding to the timeliness standard. If the platform party does not set it, the early warning duration is set with the duration corresponding to the timeliness standard; obtain the verification result data and the update waiting duration of the corresponding knowledge retrieval data in real time. If there is no corresponding verification result data currently, the verification result data is none; perform real-time screening on the knowledge retrieval data according to the verification result data, that is, eliminate the knowledge retrieval data with false and inaccurate verification result data, and also retain the data with no verification result data; Perform real-time analysis on the corresponding knowledge retrieval data according to the early warning duration and the update waiting duration. When the update waiting duration reaches the early warning duration and is not eliminated, supplement the knowledge retrieval data to the supplementary data of the supplementary analysis field.

[0031] In one embodiment, perform real-time analysis on the supplementary analysis field according to the timeliness standard, and use the existing analysis method to collect in real time the knowledge data to be supplemented in the supplementary analysis field under the premise of meeting the timeliness standard, and integrate to obtain supplementary data.

[0032] The supplementary acquisition module is used to perform real-time data acquisition based on supplementary data to obtain supplementary acquisition data in the corresponding professional field; that is, to perform standardized acquisition according to the acquisition mode, requirements, etc. of knowledge data; the supplementary acquisition data includes potential knowledge data and verification association data; the potential knowledge data is knowledge data acquired according to the corresponding knowledge retrieval data, but has not yet been supplemented and stored as knowledge data; the verification association data is real-time acquisition data for the potential knowledge data to prove whether it is real and correct association data; subsequent acquisition can be combined with the verification result data.

[0033] The trust analysis module is used to perform real-time trust analysis on the supplementary acquisition data to obtain the trust level corresponding to the corresponding supplementary acquisition data. The trust levels include untrusted level, trust qualified level, and accurate trust level. Eliminate the supplementary acquisition data with the untrusted level, mark the potential knowledge data corresponding to the remaining supplementary acquisition data as supplementary knowledge data, and mark the corresponding trust level label for the corresponding supplementary knowledge data; store the supplementary knowledge data in the knowledge base.

[0034] In one embodiment, the supplementary knowledge data stored in the knowledge base is classified and stored according to the trust level, that is, the supplementary knowledge data with the accurate trust level is preferentially applied during subsequent responses.

[0035] In one embodiment, to perform real-time trust analysis on the supplementary acquisition data, it can be analyzed based on existing trust analysis methods to determine its trust value, and the corresponding trust level is determined according to the trust value and the preset trust level standard; the trust level standard is the trust value intervals corresponding to the three trust levels, which are set by the platform side. For example, 0-59 is the untrusted level, 60-80 is the trust qualified level, and 81-100 is the accurate trust level.

[0036] Exemplarily, use machine learning algorithms (such as decision trees, random forests, support vector machines, etc.) to train a classification model, and predict the authenticity of the data according to data features (such as source, content, propagation path, etc.). Collect a large amount of labeled network data (true or false), extract features and train the model, and then predict new data to output the trust value.

[0037] Use deep learning techniques (such as convolutional neural network CNN, recurrent neural network RNN, Transformer, etc.) to process multi-modal data such as text, images, and videos, and capture complex patterns and features in the data. Build a deep learning model, input the feature representation of the network data, and output the authenticity probability or trust value of the data.

[0038] Use NLP technology to analyze the text content of network data, identify information such as semantics, sentiment, and topics, and assist in judging the authenticity of the data. Perform word segmentation, part-of-speech tagging, named entity recognition, etc. on the text, extract key information, and combine machine learning or deep learning models for authenticity assessment.

[0039] Combine the prediction results of multiple intelligent models (such as classification models, deep learning models, NLP models, etc.), and obtain the final trust value through methods such as voting and weighted average. Train multiple base models, fuse their prediction results, and improve the accuracy and stability of the assessment.

[0040] Mark the intelligent model for evaluating the above trust value as the trust value evaluation model.

[0041] In one embodiment, perform trust level update analysis on the supplementary knowledge data stored in the knowledge base; use the determination method of the trust level to perform update analysis to determine whether its trust level changes; the verification association data at this time can also include data with verification effects such as the Q&A feedback and records of the Q&A personnel.

[0042] In one embodiment, perform trust level update analysis on the supplementary knowledge data, and the trust level is the qualified trust level.

[0043] In one embodiment, perform trust level update analysis on all the supplementary knowledge data.

[0044] In one embodiment, the trust level update analysis of the supplementary knowledge data can be performed in ways such as regularly, periodically, at intervals, or in real time. For example, different analysis times can be adopted for the qualified trust level and the accurate trust level.

[0045] The Q&A module is used for users to ask questions, obtain the question data of users in real time, answer according to the knowledge base for the question data of users, obtain the corresponding answer data, and display the answer data to users.

[0046] The above formulas are all calculated by removing the dimension and taking their numerical values. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained by simulating a large amount of data.

[0047] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An artificial intelligence system for multi-field question and answer, comprising a knowledge base and a question and answer module, wherein the knowledge base is used to store knowledge data in various professional fields; the question and answer module is used for users to ask questions and answer, obtain user question data in real time, generate corresponding answer data according to the question data, and display the answer data to the user; characterized in that: It also includes an update analysis module, a supplementary acquisition module, and a trust analysis module; The update analysis module is used to perform real-time update analysis on the knowledge data of the corresponding professional field and determine the supplementary data of the corresponding professional field; The supplementary collection module is used to perform real-time data collection based on the supplementary data to obtain supplementary collected data in the corresponding professional field, wherein the supplementary collected data includes potential knowledge data and verification related data; The trust analysis module is used to perform real-time trust analysis on the supplementary collected data to obtain the trust level corresponding to the corresponding supplementary collected data, where the trust level includes an untrustworthy level, a qualified trust level, and an accurate trust level; Eliminate the supplementary collected data with an untrustworthy trust level, identify the potential knowledge data corresponding to the remaining supplementary collected data, mark the potential knowledge data as supplementary knowledge data, and mark the supplementary knowledge data with a corresponding trust level label; store the supplementary knowledge data in a knowledge base.

2. The multi-domain question-answering artificial intelligence system according to claim 1, characterized in that: Real-time update and analysis of knowledge data in corresponding professional fields, including: Preset timeliness standards for various professional fields, including qualification standards and qualification rate standards; Conduct real-time analysis on corresponding professional fields according to timeliness standards to obtain supplementary analysis fields; conduct real-time analysis on supplementary analysis fields according to timeliness standards to obtain corresponding supplementary data.

3. The multi-domain question-answering artificial intelligence system according to claim 1, characterized in that: Real-time analysis of relevant professional fields according to timeliness standards, including: A retrieval unit is provided, wherein the retrieval unit is used to retrieve in real time knowledge-related data not stored in the corresponding professional field in the knowledge base, wherein the knowledge-related data includes the retrieved knowledge data and verification result data of the corresponding retrieved knowledge data; Screening the knowledge-related data according to the verification result data to obtain domain material data of the professional field; Acquire the knowledge update records of the corresponding professional fields in the knowledge base in real time; calibrate and analyze the material data in each field according to the knowledge update records and timeliness standards to obtain the judgment result of whether the corresponding professional field meets the timeliness standards.

4. The multi-domain question-answering artificial intelligence system according to claim 3, characterized in that: Calibrate and analyze material data in various fields based on knowledge update records and timeliness standards, including: According to the knowledge update record, real-time identification of the update waiting time corresponding to each field material data; summarizing the update waiting time of each professional field to form an update time set of the professional field; Establish a domain evaluation model. The expression of the domain evaluation model is: ; Where: (U, ST) is the input data, U is the update duration set of the corresponding professional field; ST is the timeliness standard of the corresponding professional field; U→ST means that the corresponding update duration set meets the timeliness standard requirements; the output data is the field evaluation value LP(U, ST), and the field evaluation value is 1 or 0; Perform real-time analysis on the update duration set and timeliness standard of the professional field according to the field evaluation model to obtain a corresponding field evaluation value; Professional fields with a field evaluation value of 0 are marked as supplementary analysis fields.

5. The multi-domain question-answering artificial intelligence system according to claim 3, characterized in that: Real-time analysis of supplementary analytical areas according to timeliness criteria, including: Performing real-time retrieval of the supplementary analysis field through a retrieval unit to obtain knowledge retrieval data of the supplementary analysis field; setting a warning duration according to a timeliness standard; Acquire the verification result data and update waiting time of the knowledge retrieval data in real time, and screen the knowledge retrieval data in real time according to the verification result data; The remaining knowledge retrieval data is analyzed in real time according to the warning time and the update waiting time. When the update waiting time reaches the warning time, the knowledge retrieval data is supplemented to the supplementary data in the supplementary analysis field.

6. The multi-domain question-answering artificial intelligence system according to claim 1, characterized in that: The supplementary knowledge data stored in the knowledge base is classified and stored according to the trust level, and the priority of the response with the trust level of the accurate trust level is greater than the priority of the response with the trust level of the qualified trust level.

7. The multi-domain question-answering artificial intelligence system according to claim 1, characterized in that: Real-time trust analysis of supplementary collected data, including: The trust value interval corresponding to each trust level is preset, and the value range of the trust value is [0, 100]; Establishing a trust value evaluation model, wherein the trust value evaluation model is used to evaluate the trust value of the corresponding potential knowledge data; Identify the verification associated data of the supplementary collected data in real time, analyze the verification associated data through the trust value evaluation model, and obtain the trust value of the potential knowledge data corresponding to the supplementary collected data; The trust level of the potential knowledge data is determined according to the trust value and the trust value interval of the corresponding trust level.

8. The multi-domain question-answering artificial intelligence system according to claim 1, characterized in that: The trust level of the supplementary knowledge data stored in the knowledge base is updated and adjusted.

Images