Problem processing method and device based on artificial intelligence, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence and the field of digital medical treatment, and relates to a question processing method and device based on artificial intelligence, computer equipment and a storage medium. Preprocessing the question text to obtain a target question text; querying a target document conforming to a correlation relationship with the target question text from a medical knowledge base; generating first model input based on the target question text and the target document, and performing information extraction on the first model input through a target language model to obtain target information; generating a second model input based on the target question text and the target information, and performing reasoning processing on the second model input through a target language model to obtain an answer text; optimizing the answer text to obtain a target answer text; and returning the target answer text to the user. In addition, the invention also relates to a block chain technology, and the target answer text can be stored in a block chain. The processing efficiency and the answering accuracy of the medical questions and answers are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and the field of digital medicine, and in particular to artificial intelligence-based problem-solving methods, devices, computer equipment and storage media. Background Art

[0002] With the rapid development of artificial intelligence technology, local knowledge base question-answering systems based on large models have become an important means to improve the quality and efficiency of information services, especially in the medical field, where accurate and efficient information acquisition is of immeasurable value for clinical decision-making, patient education, and scientific research activities. However, when building such systems, user privacy protection has become a core issue that cannot be ignored. Although advanced large language models such as ChatGPT on the market currently have powerful natural language processing and knowledge reasoning capabilities, their high deployment costs, complexity of data privacy processing, and reliance on external data make direct application to local knowledge base question-answering systems face many challenges.

[0003] First, open source or private but limited large models are unable to cope with complex problems in the medical field. Medical data is highly professional, private, and timely, requiring the model to not only have a deep accumulation of medical knowledge, but also to be able to extract and reason information efficiently and accurately while protecting privacy. However, the currently available models are often unable to meet actual needs in terms of automatic summarization, reasoning, and personalized response capabilities in massive medical data.

[0004] Secondly, the workflow of traditional local knowledge base question-answering systems has significant flaws. Such systems usually rely on a simple search-recall mechanism, directly concatenating the retrieved relevant documents with the questions as the input of the large language model. This processing method not only increases the processing burden of the model, requiring the model to have extremely high text understanding and generation capabilities, but also easily leads to lengthy, unfocused, and inaccurate answers, which in turn affects the user experience. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose an artificial intelligence-based question processing method, apparatus, computer device and storage medium to solve the technical problem of low accuracy of answers generated by existing local knowledge base question and answer systems.

[0006] In order to solve the above technical problems, the embodiment of the present application provides a problem solving method based on artificial intelligence, which adopts the following technical solutions:

[0007] Receive question text entered by the user;

[0008] Preprocessing the question text to obtain a corresponding target question text;

[0009] Querying a target document that has a preset correlation with the target question text from a preset medical knowledge base;

[0010] Generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document;

[0011] Generate a corresponding second model input based on the target question text and the target information, and perform reasoning processing on the second model input through the target language model to obtain a corresponding answer text;

[0012] Optimizing the answer text to obtain a corresponding target answer text;

[0013] The target answer text is returned to the user.

[0014] Furthermore, the step of searching for a target document that has a preset correlation with the target question text from a preset medical knowledge base specifically includes:

[0015] Call the pre-trained document vector representation model;

[0016] Converting the question text into a corresponding question vector based on the vector representation model;

[0017] Calculating the similarity between the question vector and the vector of each document in the medical knowledge base respectively;

[0018] Sort all the documents in descending order based on similarity to obtain a corresponding document sorting list;

[0019] Filter out a target number of first documents from the document sorting list in sequence;

[0020] Performing key content extraction processing on the first document to obtain a corresponding second document;

[0021] The second document is used as the target document.

[0022] Furthermore, the step of optimizing the answer text to obtain the corresponding target answer text specifically includes:

[0023] Performing redundancy removal processing on the answer text to obtain a corresponding first answer text;

[0024] Performing grammatical correction processing on the first answer text to obtain a corresponding second answer text;

[0025] The second answer text is used as the target answer text.

[0026] Furthermore, the step of returning the target answer text to the user specifically includes:

[0027] Get the target display format;

[0028] Convert the target answer text based on the target display format to obtain a converted target answer text;

[0029] Call the preset user interface;

[0030] The target answer text is presented to the user based on the user interface.

[0031] Furthermore, the step of preprocessing the question text to obtain the corresponding target question text specifically includes:

[0032] Removing irrelevant information from the question text to obtain a corresponding first question text;

[0033] Get the preset standard format;

[0034] Performing format conversion processing on the first question text based on the standard format to obtain a corresponding second question text;

[0035] The second question text is used as the target question text.

[0036] Furthermore, before the step of generating a corresponding first model input based on the target question text and the target document, and extracting information from the first model input through a preset target language model to extract target information related to the question text from the target document, the method further includes:

[0037] Build medical question-answer pairs based on a preset large language model;

[0038] Annotating the medical question-answer pairs to obtain corresponding medical sample data;

[0039] Dividing the medical sample data into a training set and a test set;

[0040] Call the preset language model;

[0041] Get the preset improved cross entropy loss function;

[0042] Fine-tune the language model using the improved cross entropy loss function and the training set to obtain a corresponding specified language model;

[0043] Performing a performance test on the specified language model based on the test set;

[0044] If the specified language model passes the performance test, the specified language model is used as the target language model.

[0045] Furthermore, the step of obtaining a preset improved cross entropy loss function specifically includes:

[0046] Get the cross entropy loss function;

[0047] Get the preset weight adjustment strategy;

[0048] Adjust the cross entropy loss function based on the weight adjustment strategy to obtain an adjusted cross entropy loss function;

[0049] The adjusted cross entropy loss function is used as the improved cross entropy loss function.

[0050] In order to solve the above technical problems, the embodiment of the present application further provides a problem processing device based on artificial intelligence, which adopts the following technical solution:

[0051] A receiving module, used for receiving question text input by the user;

[0052] A preprocessing module, used for preprocessing the question text to obtain a corresponding target question text;

[0053] A query module, used to query a target document that has a preset correlation with the target question text from a preset medical knowledge base;

[0054] An extraction module, configured to generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document;

[0055] An inference module, configured to generate a corresponding second model input based on the target question text and the target information, and perform inference processing on the second model input through the target language model to obtain a corresponding answer text;

[0056] An optimization module, used for optimizing the answer text to obtain a corresponding target answer text;

[0057] A returning module is used to return the target answer text to the user.

[0058] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0059] Receive question text entered by the user;

[0060] Preprocessing the question text to obtain a corresponding target question text;

[0061] Querying a target document that has a preset correlation with the target question text from a preset medical knowledge base;

[0062] Generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document;

[0063] Generate a corresponding second model input based on the target question text and the target information, and perform reasoning processing on the second model input through the target language model to obtain a corresponding answer text;

[0064] Optimizing the answer text to obtain a corresponding target answer text;

[0065] The target answer text is returned to the user.

[0066] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0067] Receive question text entered by the user;

[0068] Preprocessing the question text to obtain a corresponding target question text;

[0069] Querying a target document that has a preset correlation with the target question text from a preset medical knowledge base;

[0070] Generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document;

[0071] Generate a corresponding second model input based on the target question text and the target information, and perform reasoning processing on the second model input through the target language model to obtain a corresponding answer text;

[0072] Optimizing the answer text to obtain a corresponding target answer text;

[0073] The target answer text is returned to the user.

[0074] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0075] The present application first receives a question text input by a user; and pre-processes the question text to obtain a corresponding target question text; then queries a target document that has a preset correlation with the target question text from a preset medical knowledge base; then generates a corresponding first model input based on the target question text and the target document, and extracts information from the first model input through a preset target language model to extract target information related to the question text from the target document; subsequently generates a corresponding second model input based on the target question text and the target information, and performs inference processing on the second model input through the target language model to obtain a corresponding answer text; further optimizes the answer text to obtain a corresponding target answer text; and finally returns the target answer text to the user. The present application constructs a local knowledge question-and-answer system for the medical field based on a preset target language model. By using the local knowledge question-and-answer system, the question text input by the user can be accurately answered, thereby improving the processing efficiency of medical questions and answers, and improving the accuracy of the generated target answer text. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0077] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0078] Figure 2 A flowchart of an embodiment of the problem-solving method based on artificial intelligence according to the present application;

[0079] Figure 3 is a structural schematic diagram of an embodiment of an artificial intelligence-based problem processing device according to the present application;

[0080] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0081] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0082] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0083] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0084] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0085] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0086] The terminal devices 101, 102, and 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, and the like.

[0087] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0088] It should be noted that the artificial intelligence-based problem handling method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based problem handling device is generally set in the server / terminal device.

[0089] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0090] Continue to refer Figure 2 , showing a flow chart of an embodiment of an artificial intelligence-based problem-solving method according to the present application. According to different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted. The artificial intelligence-based problem-solving method provided in the embodiment of the present application can be applied to any scenario that requires medical knowledge questions and answers, and the artificial intelligence-based problem-solving method can be applied to products in these scenarios, for example, medical knowledge questions and answers in the field of digital medicine. The artificial intelligence-based problem-solving method comprises the following steps:

[0091] Step S201, receiving a question text input by a user.

[0092] In this embodiment, the problem-solving method based on artificial intelligence runs on the electronic device (e.g. Figure 1The server / terminal device shown in the figure) can obtain the question text through a wired connection or a wireless connection. It should be noted that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, Wi-Fi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (Ultra Wi DeBand) connection, and other wireless connection methods currently known or developed in the future. The present application can be applied to the business scenario of medical knowledge question and answer in the field of digital medicine, and the execution subject of the present application may be a medical knowledge question and answer system. Among them, a user-friendly interface is pre-designed in the medical knowledge question and answer system to allow the user to enter questions. And the question text entered by the user in the above-mentioned user-friendly interface is received through a relevant interface or a front-end form submission mechanism. The above-mentioned question text is specifically a question related to medical care, for example: "One bar on the pregnancy test stick is very obvious and one is very shallow. Is it pregnant?".

[0093] Step S202, preprocessing the question text to obtain a corresponding target question text.

[0094] In this embodiment, the specific implementation process of preprocessing the question text to obtain the corresponding target question text will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0095] Step S203, searching the preset medical knowledge base for target documents that have a preset correlation with the target question text.

[0096] In this embodiment, the above-mentioned medical knowledge base is a local medical knowledge base. The process of building a medical knowledge base includes: Collecting data: Collecting medical-related documents from various sources (such as medical books, research papers, medical websites, government reports, etc.). Cleaning and sorting: Cleaning the collected documents, including removing irrelevant information, unifying the format, etc. Storage: Store the cleaned and sorted documents in a local database to ensure quick access. Among them, the specific implementation process of querying the target document that meets the preset correlation relationship with the target problem text from the preset medical knowledge base will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated on here.

[0097] Step S204, generating a corresponding first model input based on the target question text and the target document, and performing information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document.

[0098] In this embodiment, the target question text and the target document are spliced ​​into a format that can be understood by the target language model as the first model input. Then, the target language model is used to extract information from the first model input through an extractive method, extract information related to the target question text, and organize the extracted information into a structured form (such as JSON, XML, etc.) to obtain the target information. Among them, the model construction process of the target language model will be further described in detail in the subsequent specific embodiments, and will not be elaborated on here.

[0099] Step S205: Generate a corresponding second model input based on the target question text and the target information, and perform inference processing on the second model input through the target language model to obtain a corresponding answer text.

[0100] In this embodiment, the target question text and the target information are spliced ​​into a format that the target language model can understand as the second model input. The target language model is then used again to generate a coherent and accurate answer text based on the input information in a generative manner.

[0101] Step S206, optimizing the answer text to obtain a corresponding target answer text.

[0102] In this embodiment, the specific implementation process of optimizing the answer text to obtain the corresponding target answer text will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0103] Step S207: returning the target answer text to the user.

[0104] In this embodiment, the specific implementation process of returning the target answer text to the user will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0105] The present application first receives a question text input by a user; and pre-processes the question text to obtain a corresponding target question text; then queries a target document that has a preset correlation with the target question text from a preset medical knowledge base; then generates a corresponding first model input based on the target question text and the target document, and extracts information from the first model input through a preset target language model to extract target information related to the question text from the target document; subsequently generates a corresponding second model input based on the target question text and the target information, and performs inference processing on the second model input through the target language model to obtain a corresponding answer text; further optimizes the answer text to obtain a corresponding target answer text; and finally returns the target answer text to the user. The present application constructs a local knowledge question-and-answer system for the medical field based on a preset target language model. By using the local knowledge question-and-answer system, the question text input by the user can be accurately answered, thereby improving the processing efficiency of medical questions and answers, and improving the accuracy of the generated target answer text.

[0106] In some optional implementations, step S203 includes the following steps:

[0107] Call the pre-trained document vector representation model.

[0108] In this embodiment, a part of documents can be selected from the medical knowledge base as a training data set, and a pre-selected initial model can be trained to construct a document vector representation model having a model function of converting documents into high-dimensional vectors. The above initial model adopts a Transformer-based model, such as BERT, RoBERTa, etc.

[0109] The question text is converted into a corresponding question vector based on the vector representation model.

[0110] In this embodiment, the question text is input into the vector representation model so as to convert the question text into a corresponding question vector through the vector representation model.

[0111] The similarity between the question vector and the vector of each document in the medical knowledge base is calculated respectively.

[0112] In this embodiment, the similarity between the question vector and the vector of each document in the medical knowledge base can be calculated by using a similarity algorithm. There is no specific limitation on the selection of the similarity algorithm, for example, cosine similarity, Euclidean distance, etc. can be used.

[0113] All the documents are sorted in descending order based on similarity to obtain a corresponding document sorting list.

[0114] In this embodiment, all the documents may be sorted in descending order according to similarity, and the document with the highest similarity score is most relevant to the question input by the user.

[0115] A target number of first documents are sequentially screened out from the document sorting list.

[0116] In this embodiment, the first m most relevant documents (doc_1, doc_2, ..., doc_m) can be selected from the sorted document ranking list as the first document. The value of the target number m is not specifically limited and can be set according to actual business usage requirements, for example, it can be set to 5.

[0117] Key content extraction is performed on the first document to obtain a corresponding second document.

[0118] In this embodiment, the key content extraction process is performed on the first document to extract the key paragraphs in the first document, thereby obtaining the above-mentioned second document, so that the data processing workload can be reduced during subsequent information extraction, thereby improving the extraction efficiency of information extraction.

[0119] The second document is used as the target document.

[0120] This application calls a pre-trained document vector representation model; and based on the vector representation model, converts the question text into a corresponding question vector; then respectively calculates the similarity between the question vector and the vector of each document in the medical knowledge base; then sorts all the documents in descending order based on the similarity to obtain a corresponding document sorting list; subsequently, sequentially selects a target number of first documents from the document sorting list; further performs key content extraction processing on the first document to obtain a corresponding second document; and finally uses the second document as the target document. This application performs document query processing on the medical knowledge base with respect to the target question text based on the use of a document vector representation model and a processing method for vector similarity calculation, which can realize fast and accurate querying of the target document most relevant to the target question text from the medical knowledge base, improves the query efficiency of the target document, and ensures the data accuracy of the obtained target document.

[0121] In some optional implementations of this embodiment, step S206 includes the following steps:

[0122] The answer text is processed to remove redundancy to obtain a corresponding first answer text.

[0123] In this embodiment, irrelevant contents, such as modal particles, stop words, etc., are removed from the answer text to complete redundancy removal processing and obtain the corresponding first answer text.

[0124] The first answer text is subjected to grammatical correction processing to obtain a corresponding second answer text.

[0125] In this embodiment, the grammar correction processing refers to the processing of correcting grammatical errors in the first answer text, thereby ensuring the grammatical accuracy of the obtained second answer text.

[0126] The second answer text is used as the target answer text.

[0127] This application obtains the corresponding first answer text by removing redundancy from the answer text; then corrects the grammar of the first answer text to obtain the corresponding second answer text; and subsequently uses the second answer text as the target answer text. After the application obtains the corresponding answer text by inference processing the second model input through the target language model, it further removes redundancy and corrects the grammar of the answer text to obtain the corresponding target answer text, effectively ensuring the conciseness and accuracy of the target answer text. This is beneficial for the subsequent return of the target answer text to the user, which can ensure the user's browsing experience.

[0128] In some optional implementations, step S207 includes the following steps:

[0129] Get the target display format.

[0130] In this embodiment, there is no specific limitation on the selection of the above-mentioned target display mode, which may be a mode suitable for display, such as HTML, Markdown, etc.

[0131] The target answer text is converted based on the target display format to obtain a converted target answer text.

[0132] In this embodiment, the target answer text is converted into a format that matches the target display format, thereby obtaining the converted target answer text.

[0133] Call the default user interface.

[0134] In this embodiment, the above user interface is a pre-built interface for data interaction with the user.

[0135] The target answer text is presented to the user based on the user interface.

[0136] In this embodiment, the target answer text is displayed to the user by using the above-mentioned user interface, for example, it may include directly displaying it in the client, or notifying the user through email, text message, etc.

[0137] The present application obtains a target display format; then converts the target answer text based on the target display format to obtain a converted target answer text; then calls a preset user interface; and subsequently displays the target answer text to the user based on the user interface. After optimizing the answer text to obtain the corresponding target answer text, the present application will also intelligently convert the target answer text based on the target display format to obtain a converted target answer text, so as to convert the target answer text into a format suitable for display. Subsequently, the target answer text is displayed to the user based on a preset user interface, so that the user can comfortably view the target answer text, thereby improving the user experience and the display intelligence of the target answer text.

[0138] In some optional implementations, step S202 includes the following steps:

[0139] The question text is processed to remove irrelevant information to obtain a corresponding first question text.

[0140] In this embodiment, the irrelevant information may specifically include irrelevant symbols.

[0141] Gets the preset standard format.

[0142] In this embodiment, the standard format may specifically be a pre-set text standard format, for example, a unified lowercase format or a unified uppercase format.

[0143] The first question text is format converted based on the standard format to obtain a corresponding second question text.

[0144] In this embodiment, the corresponding second question text is obtained by converting the first question text into a format matching the above standard format.

[0145] The second question text is used as the target question text.

[0146] The present application removes irrelevant information from the question text to obtain a corresponding first question text; then obtains a preset standard format; then performs format conversion on the first question text based on the standard format to obtain a corresponding second question text; and subsequently uses the second question text as the target question text. The present application removes irrelevant information from the question text and performs format conversion based on the standard format, thereby achieving fast and intelligent preprocessing of the question text and ensuring the data accuracy of the obtained target question text.

[0147] In some optional implementations of this embodiment, before step S204, the electronic device may further perform the following steps:

[0148] Build medical question-answer pairs based on a preset large language model.

[0149] In this embodiment, the above-mentioned large language model can specifically adopt a large language model such as ChatGPT. The process of constructing a medical question-answer pair includes: designing a series of medical-related question templates according to actual medical treatment needs, or directly using the ChatGPT model to automatically generate medical questions. Then, the above-mentioned medical questions are answered using a basic data set or an external knowledge base (such as an online medical forum, a question-and-answer community) to generate corresponding medical question-answer pairs.

[0150] The medical question and answer pairs are labeled to obtain corresponding medical sample data.

[0151] In this embodiment, medical experts or trained data annotators can manually review and annotate the generated medical question and answer pairs, and ensure the accuracy, relevance and rationality of the medical question and answer pairs, so as to obtain the corresponding initial medical sample data. In addition, the annotated initial medical sample data is organized into a format suitable for model training to obtain the above-mentioned medical sample data.

[0152] The medical sample data is divided into a training set and a test set.

[0153] In this embodiment, the medical sample data may be divided into a training set and a test set according to a preset division ratio. There is no specific limitation on the numerical value of the division ratio, for example, 7:3 may be used.

[0154] Call the preset language model.

[0155] In this embodiment, the above-mentioned language model can specifically adopt language models such as BERT, GPT, Transformer, etc. And according to the actual model construction requirements, set the hyperparameters of the language model such as learning rate, batch size, training rounds, etc. And configure the optimizer (such as Adam, AdamW) and scheduler (such as Warmup-Cos i ne Annea ling). Among them, fine-tuning usually uses a smaller data set, then a smaller batch size and more training rounds are used. In addition, the model architecture can be adjusted as needed, such as adding task-specific layers (such as classification layers, decoding layers, etc.) to support medical question-answering tasks.

[0156] Get the preset improved cross entropy loss function.

[0157] In this embodiment, the specific implementation process of obtaining the preset improved cross entropy loss function will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated in detail here.

[0158] The language model is fine-tuned using the improved cross entropy loss function and the training set to obtain a corresponding specified language model.

[0159] In this embodiment, the language model is trained to learn manually annotated answers to achieve better generation results. This step uses the manually annotated answers as the standard, and the training goal is to increase the probability of the language model generating this sentence, or to maximize the log-likelihood function of this reply under the language model. Specifically, the language model is fine-tuned by using a training set and an improved cross-entropy loss function, and the loss value, accuracy and other indicators are monitored during the training process. The language model parameters are adjusted or the model is optimized as needed, and strategies such as early stopping are used to prevent overfitting, and then the corresponding specified language model is trained.

[0160] A performance test is performed on the specified language model based on the test set.

[0161] In this embodiment, the test set is input into the specified language model for testing, and the model performance (question and answer accuracy, generation quality) of the fine-tuned specified language model is evaluated on the test set. If the obtained question and answer accuracy is greater than a preset accuracy threshold and the generation quality meets the preset quality standard, the specified language model is determined to have passed the performance test; otherwise, the specified language model is determined to have failed the performance test.

[0162] If the specified language model passes the performance test, the specified language model is used as the target language model.

[0163] In this embodiment, if the specified language model passes the performance test, it is determined that the prediction capability of the specified language model meets the construction requirements, and the specified language model is used as the target language model. Subsequently, the trained target language model is deployed to the production environment for use, and the model performance of the target language model is monitored in real time, user feedback is collected, and continuous optimization is performed.

[0164] This application constructs a medical question-and-answer pair based on a preset large language model; then annotates the medical question-and-answer pair to obtain corresponding medical sample data, and divides the medical sample data into a training set and a test set; then calls the preset language model; and obtains a preset improved cross-entropy loss function; subsequently uses the improved cross-entropy loss function and the training set to fine-tune the language model to obtain the corresponding designated language model; further performs a performance test on the designated language model based on the test set; if the designated language model passes the performance test, the designated language model is used as the target language model. This application constructs a medical question-and-answer pair based on a preset large language model, and annotates the medical question-and-answer pair to obtain corresponding medical sample data, and then uses the improved cross-entropy loss function and the medical sample data to fine-tune and test the preset language model, thereby realizing the rapid and intelligent construction of a target language model that meets the requirements, improving the construction efficiency of the target language model, and ensuring the model effect of the target language model obtained. In addition, by fine-tuning the language model using the improved cross-entropy loss function, the overfitting phenomenon of the language model can be alleviated during the training of the language model, the training effect of the language model can be enhanced, and the performance of the language model in question-answering tasks can be improved.

[0165] In some optional implementations of this embodiment, obtaining a preset improved cross entropy loss function includes the following steps:

[0166] Get the cross entropy loss function.

[0167] In this embodiment, the cross entropy loss function specifically includes: L = ∑t i log(p i ). Where L is the cross entropy loss function, t i The value is 0 or 1. If it is 1, it means the true label is equal to i, p i is the probability predicted by the model for label i.

[0168] Get the preset weight adjustment strategy.

[0169] In this embodiment, the cross entropy loss function has the same weight for each token, which can easily lead to the following situation: the model is quickly fitted on simple token predictions, which quickly reduces the level of the loss function to a very low level and falls into a local optimal solution. However, the prediction results of the model on difficult tokens are still very poor. Therefore, the following improvement is proposed in this application, that is, the content of the weight adjustment strategy includes: for each predicted token i, the probability distribution of the model prediction is now P i , by calculating the entropy of the probability distribution: E i =Entropy(P i ). For a distribution with larger entropy, the uncertainty is greater, which means that the model has no confidence in this prediction, so it is necessary to give a larger weight to the loss of this token: β i =*( i )+b, where a is a positive number that controls the effect of entropy on weights. b is a constant (used to ensure that weights are not zero even if entropy is small).

[0170] The cross entropy loss function is adjusted based on the weight adjustment strategy to obtain an adjusted cross entropy loss function.

[0171] In this embodiment, the cross entropy loss function can be adjusted according to the strategy content of the weight adjustment strategy to obtain an adjusted cross entropy loss function. Specifically, the adjusted cross entropy loss function, that is, the improved cross entropy loss function, specifically includes: L = ∑ β i *t i *log(p i ). Among them, L is the improved cross entropy loss function, β i The weight of each predicted token, t i The value is 0 or 1. If it is 1, it means the true label is equal to i, p i is the probability predicted by the model for label i.

[0172] The adjusted cross entropy loss function is used as the improved cross entropy loss function.

[0173] In this embodiment, an improved cross entropy loss function is introduced to weight the loss of each token according to its prediction uncertainty to improve the prediction ability of the model on difficult tokens.

[0174] The present application obtains a cross entropy loss function; then obtains a preset weight adjustment strategy; then adjusts the cross entropy loss function based on the weight adjustment strategy to obtain an adjusted cross entropy loss function; and subsequently uses the adjusted cross entropy loss function as the improved cross entropy loss function. The present application adjusts the cross entropy loss function using a preset weight adjustment strategy, thereby intelligently generating the required improved cross entropy loss function, so that when the improved cross entropy loss function and the training set are subsequently used to fine-tune the language model, the overfitting phenomenon of the language model can be alleviated during the training of the language model, the training effect of the language model can be enhanced, and the performance of the language model in question-answering tasks can be improved.

[0175] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0176] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned target answer text, the above-mentioned target answer text can also be stored in a node of a blockchain.

[0177] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0178] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0179] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0180] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0181] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0182] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a problem-solving device based on artificial intelligence. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0183] like Figure 3 As shown, the problem processing device 300 based on artificial intelligence described in this embodiment includes: a receiving module 301, a preprocessing module 302, a query module 303, an extraction module 304, a reasoning module 305, an optimization module 306 and a return module 307. Among them:

[0184] Receiving module 301, used for receiving question text input by a user;

[0185] A preprocessing module 302 is used to preprocess the question text to obtain a corresponding target question text;

[0186] A query module 303 is used to query a target document that has a preset correlation with the target question text from a preset medical knowledge base;

[0187] An extraction module 304 is used to generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document;

[0188] An inference module 305 is used to generate a corresponding second model input based on the target question text and the target information, and to perform inference processing on the second model input through the target language model to obtain a corresponding answer text;

[0189] An optimization module 306 is used to optimize the answer text to obtain a corresponding target answer text;

[0190] The return module 307 is used to return the target answer text to the user.

[0191] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0192] In some optional implementations of this embodiment, the query module 303 includes:

[0193] The first calling submodule is used to call a pre-trained document vector representation model;

[0194] A first conversion submodule, configured to convert the question text into a corresponding question vector based on the vector representation model;

[0195] A calculation submodule, used for respectively calculating the similarity between the question vector and the vector of each document in the medical knowledge base;

[0196] A sorting submodule is used to sort all the documents in descending order based on similarity to obtain a corresponding document sorting list;

[0197] A screening submodule, used to sequentially screen out a target number of first documents from the document sorting list;

[0198] An extraction submodule, used to extract key content from the first document to obtain a corresponding second document;

[0199] The first determination submodule is used to take the second document as the target document.

[0200] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0201] In some optional implementations of this embodiment, the optimization module 306 includes:

[0202] A first processing submodule is used to remove redundancy from the answer text to obtain a corresponding first answer text;

[0203] A second processing submodule is used to perform grammar correction processing on the first answer text to obtain a corresponding second answer text;

[0204] The second determination submodule is used to use the second answer text as the target answer text.

[0205] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0206] In some optional implementations of this embodiment, the returning module 307 includes:

[0207] A first acquisition submodule is used to acquire a target display format;

[0208] A second conversion submodule is used to convert the target answer text based on the target display format to obtain a converted target answer text;

[0209] The second calling submodule is used to call a preset user interface;

[0210] A display submodule is used to display the target answer text to the user based on the user interface.

[0211] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0212] In some optional implementations of this embodiment, the preprocessing module 302 includes:

[0213] A third processing submodule is used to remove irrelevant information from the question text to obtain a corresponding first question text;

[0214] The second acquisition submodule is used to acquire a preset standard format;

[0215] A third conversion submodule is used to perform format conversion processing on the first question text based on the standard format to obtain a corresponding second question text;

[0216] The third determination submodule is used to use the second question text as the target question text.

[0217] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0218] In some optional implementations of this embodiment, the problem processing device based on artificial intelligence further includes:

[0219] A building module for constructing medical question-answer pairs based on a preset large language model;

[0220] A labeling module, used to label the medical question and answer pairs to obtain corresponding medical sample data;

[0221] A division module, used for dividing the medical sample data into a training set and a test set;

[0222] A calling module is used to call a preset language model;

[0223] An acquisition module is used to obtain a preset improved cross entropy loss function;

[0224] A fine-tuning module, used to fine-tune the language model using the improved cross entropy loss function and the training set to obtain a corresponding specified language model;

[0225] A testing module, used for performing a performance test on the specified language model based on the test set;

[0226] The determination module is configured to use the specified language model as the target language model if the specified language model passes the performance test.

[0227] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0228] In some optional implementations of this embodiment, the acquisition module includes:

[0229] The third acquisition submodule is used to obtain the cross entropy loss function;

[0230] The fourth acquisition submodule is used to acquire a preset weight adjustment strategy;

[0231] An adjustment submodule, used to adjust the cross entropy loss function based on the weight adjustment strategy to obtain an adjusted cross entropy loss function;

[0232] The fourth determination submodule is used to use the adjusted cross entropy loss function as the improved cross entropy loss function.

[0233] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the artificial intelligence-based problem-solving method in the aforementioned implementation mode, and will not be repeated here.

[0234] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0235] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital processors (DSPs), embedded devices, etc.

[0236] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0237] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of a problem-solving method based on artificial intelligence, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0238] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the computer-readable instructions stored in the memory 41 or process data, such as running the computer-readable instructions of the problem-solving method based on artificial intelligence.

[0239] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0240] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0241] In an embodiment of the present application, firstly, a question text input by a user is received; and the question text is preprocessed to obtain a corresponding target question text; then, a target document that meets a preset correlation with the target question text is queried from a preset medical knowledge base; then, a corresponding first model input is generated based on the target question text and the target document, and information is extracted from the first model input through a preset target language model to extract target information related to the question text from the target document; subsequently, a corresponding second model input is generated based on the target question text and the target information, and the second model input is inferred through the target language model to obtain a corresponding answer text; the answer text is further optimized to obtain a corresponding target answer text; and finally, the target answer text is returned to the user. The present application constructs a local knowledge question-and-answer system for the medical field based on a preset target language model. By using the local knowledge question-and-answer system, the question text input by the user can be accurately answered, thereby improving the processing efficiency of medical questions and answers, and improving the accuracy of the generated target answer text.

[0242] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the artificial intelligence-based problem-solving method as described above.

[0243] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0244] In an embodiment of the present application, firstly, a question text input by a user is received; and the question text is preprocessed to obtain a corresponding target question text; then, a target document that meets a preset correlation with the target question text is queried from a preset medical knowledge base; then, a corresponding first model input is generated based on the target question text and the target document, and information is extracted from the first model input through a preset target language model to extract target information related to the question text from the target document; subsequently, a corresponding second model input is generated based on the target question text and the target information, and the second model input is inferred through the target language model to obtain a corresponding answer text; the answer text is further optimized to obtain a corresponding target answer text; and finally, the target answer text is returned to the user. The present application constructs a local knowledge question-and-answer system for the medical field based on a preset target language model. By using the local knowledge question-and-answer system, the question text input by the user can be accurately answered, thereby improving the processing efficiency of medical questions and answers, and improving the accuracy of the generated target answer text.

[0245] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0246] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A problem-solving method based on artificial intelligence, characterized in that: The steps include: Receive question text entered by the user; Preprocessing the question text to obtain a corresponding target question text; Querying a target document that has a preset correlation with the target question text from a preset medical knowledge base; Generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document; Generate a corresponding second model input based on the target question text and the target information, and perform reasoning processing on the second model input through the target language model to obtain a corresponding answer text; Optimizing the answer text to obtain a corresponding target answer text; The target answer text is returned to the user.

2. The problem-solving method based on artificial intelligence according to claim 1, characterized in that: The step of searching for a target document that has a preset correlation with the target question text from a preset medical knowledge base specifically includes: Call the pre-trained document vector representation model; Converting the question text into a corresponding question vector based on the vector representation model; Calculating the similarity between the question vector and the vector of each document in the medical knowledge base respectively; Sort all the documents in descending order based on similarity to obtain a corresponding document sorting list; Filter out a target number of first documents from the document sorting list in sequence; Performing key content extraction processing on the first document to obtain a corresponding second document; The second document is used as the target document.

3. The problem-solving method based on artificial intelligence according to claim 1, characterized in that: The step of optimizing the answer text to obtain the corresponding target answer text specifically includes: Performing redundancy removal processing on the answer text to obtain a corresponding first answer text; Performing grammatical correction processing on the first answer text to obtain a corresponding second answer text; The second answer text is used as the target answer text.

4. The problem-solving method based on artificial intelligence according to claim 1, characterized in that: The step of returning the target answer text to the user specifically includes: Get the target display format; Convert the target answer text based on the target display format to obtain a converted target answer text; Call the preset user interface; The target answer text is presented to the user based on the user interface.

5. The problem-solving method based on artificial intelligence according to claim 1, characterized in that: The step of preprocessing the question text to obtain the corresponding target question text specifically includes: Removing irrelevant information from the question text to obtain a corresponding first question text; Get the preset standard format; Performing format conversion processing on the first question text based on the standard format to obtain a corresponding second question text; The second question text is used as the target question text.

6. The problem-solving method based on artificial intelligence according to claim 1, characterized in that: Before the step of generating a corresponding first model input based on the target question text and the target document, and extracting information from the first model input through a preset target language model to extract target information related to the question text from the target document, the method further includes: Build medical question-answer pairs based on a preset large language model; Annotating the medical question-answer pairs to obtain corresponding medical sample data; Dividing the medical sample data into a training set and a test set; Call the preset language model; Get the preset improved cross entropy loss function; Fine-tune the language model using the improved cross entropy loss function and the training set to obtain a corresponding specified language model; Performing a performance test on the specified language model based on the test set; If the specified language model passes the performance test, the specified language model is used as the target language model.

7. The problem-solving method based on artificial intelligence according to claim 6, characterized in that: The step of obtaining a preset improved cross entropy loss function specifically includes: Get the cross entropy loss function; Get the preset weight adjustment strategy; Adjust the cross entropy loss function based on the weight adjustment strategy to obtain an adjusted cross entropy loss function; The adjusted cross entropy loss function is used as the improved cross entropy loss function.

8. A problem-solving device based on artificial intelligence, characterized in that: include: A receiving module, used for receiving question text input by the user; A preprocessing module, used for preprocessing the question text to obtain a corresponding target question text; A query module, used to query a target document that has a preset correlation with the target question text from a preset medical knowledge base; An extraction module, configured to generate a corresponding first model input based on the target question text and the target document, and perform information extraction on the first model input through a preset target language model to extract target information related to the question text from the target document; An inference module, configured to generate a corresponding second model input based on the target question text and the target information, and perform inference processing on the second model input through the target language model to obtain a corresponding answer text; An optimization module, used for optimizing the answer text to obtain a corresponding target answer text; A returning module is used to return the target answer text to the user.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the problem-solving method based on artificial intelligence as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the artificial intelligence-based problem-solving method according to any one of claims 1 to 7.

Images