Professional intelligent question-answering system and method for oil and gas geological survey and system thereof

Abstract

The invention discloses a professional intelligent question answering system and method for oil and gas geological survey, and the method comprises the steps: receiving a question inputted by a user, and extracting a professional vocabulary from the inputted question; interpreting and translating the extracted professional vocabularies by using a translation model to form popular interpretation about the professional vocabularies; constructing a knowledge base in the field of oil and gas geology survey, extracting keywords in popular interpretations of professional vocabularies, and matching related documents from the knowledge base by taking the keywords as retrieval conditions; key information in the related documents is extracted, and the extracted key information is fed back to the user through the front-end interface; the system provided by the invention can automatically identify the professional vocabularies in the questions, and explains the professional vocabularies to expand the matching rate and the matching accuracy of the output matching answer content, so that seamless user experience is provided, and both professional users and non-professional users can obtain accurate information.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas geological survey question and answer systems, and in particular to a professional intelligent question and answer system and method and system thereof for oil and gas geological survey. Background Art

[0002] In oil and gas geological survey, professional terminology is a highly specialized and technical field, involving a large number of professional words and terms, which are difficult for non-professionals to understand, affecting the efficiency of interdisciplinary communication. With the increasing frequency of international exchanges, the accurate translation and analysis of these professional terms is particularly important. Therefore, the development of professional terminology translation and analysis methods and technologies for oil and gas geological survey is imminent, which can improve translation efficiency while ensuring accuracy.

[0003] In the field of oil and gas geological survey, there are currently a variety of technical means to support the work of professionals, especially in information query and document management. Existing solutions mainly include:

[0004] 1. Direct dialogue with general language model + knowledge base query: After the user enters a query request, the system directly communicates through a general language model (such as BERT, GPT, etc.), extracts the keywords entered by the user, and then searches for relevant documents from the knowledge base through keyword matching.

[0005] 2. Rule matching + keyword search: Extract keywords from the questions input by users through predefined rules and keyword lists, and then search for relevant documents from the knowledge base through keyword matching. It is suitable for query tasks with clear rules and fixed keywords.

[0006] Therefore, although the above scheme has met the needs of oil and gas geological survey to a certain extent, it still has some shortcomings, which are mainly reflected in the following aspects:

[0007] (1) Low knowledge base matching rate: The field of oil and gas geological survey involves a large number of professional terms, which often have specific meanings and usages. For some non-professional users, they may enter imprecise keywords, and conventional language models are difficult to accurately fine-tune the keywords to form professional vocabulary. If the professional vocabulary recognition is inaccurate, subsequent translation and analysis will produce errors, affecting the overall performance of the system.

[0008] (2) Poor translation quality of professional vocabulary: Due to the high degree of professionalism in the field of oil and gas geological survey, many key words need further explanation before they can be converted into a form that is easy for ordinary users to understand. Even if professional vocabulary is identified, existing systems often lack a "translation" mechanism for professional vocabulary and may not be able to accurately translate these professional vocabulary into easy-to-understand language. Moreover, professional vocabulary often has multiple meanings and is prone to ambiguity during the translation process.

[0009] (3) Poor user experience: For non-professionals and some professionals, even if the system can provide relevant literature links or abstracts, the lack of explanation and translation of professional terms affects the effect of information acquisition. Summary of the invention

[0010] The purpose of the present invention is to provide a professional intelligent question-answering system and method and system for oil and gas geological surveys, so as to solve the technical problems in the prior art that it is difficult to accurately identify non-professional keywords, lack a "translation" mechanism for professional vocabulary, and may not be able to accurately translate these professional vocabulary into easy-to-understand language. Due to the lack of explanation and translation of professional terms, users may still find it difficult to understand the core content of the information, affecting the effect of information acquisition.

[0011] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0012] A professional intelligent question-answering method for oil and gas geological survey, comprising the following steps:

[0013] Step 100: receiving a question input by a user, and extracting professional vocabulary from the input question;

[0014] Step 200: using a translation model to interpret and translate the extracted professional vocabulary to form a popular interpretation of the professional vocabulary;

[0015] Step 300: construct a knowledge base in the field of oil and gas geological survey, extract keywords from popular explanations of professional terms, and use the keywords as search conditions to match relevant documents from the knowledge base;

[0016] Step 400: extract key information from relevant documents, and feed back the extracted key information to the user through the front-end interface.

[0017] As a preferred solution of the present invention, in step 100, the method for extracting professional vocabulary from the input question is:

[0018] Collect relevant documents of oil and gas geological survey to establish a professional vocabulary database;

[0019] Capture the contextual information and semantic features of the questions entered by the user, and combine with the customized rule base to identify the professional vocabulary in the questions entered by the user.

[0020] As a preferred solution of the present invention, the implementation steps of establishing a professional vocabulary database are:

[0021] Collect a large amount of documents related to oil and gas geological surveys, analyze and process the documents, and build a professional vocabulary database containing special terms and terminology in the field of oil and gas geology;

[0022] The explanation and translation content corresponding to each professional vocabulary is integrated into the professional vocabulary library, and the professional vocabulary in the professional vocabulary library is annotated, so that the professional vocabulary with consistent annotations is retained in the professional vocabulary library, and the professional vocabulary with inconsistent annotations is removed.

[0023] As a preferred solution of the present invention, the steps for implementing the capture and recognition of professional vocabulary in the question input by the user in combination with the professional vocabulary library are as follows:

[0024] Perform word segmentation on the questions input by the user, breaking the questions input by the user into multiple independent words or phrases;

[0025] Create a stop word list to filter out stop words in questions entered by users;

[0026] Perform part-of-speech tagging on the results after word segmentation to distinguish the decomposed words into nouns, verbs or adjectives;

[0027] Matching the words after word segmentation and part-of-speech tagging with a pre-established professional vocabulary database, and extracting the same professional vocabulary as that in the professional vocabulary database from the words after word segmentation and part-of-speech tagging;

[0028] Combined with context analysis, two adjacent professional vocabulary are further combined to form an overall vocabulary, and the overall vocabulary is matched with a pre-established professional vocabulary library;

[0029] The custom rule base further filters the extracted professional vocabulary that is the same as that in the professional vocabulary base to exclude the vocabulary that belongs to the professional vocabulary but is irrelevant to the current problem.

[0030] As a preferred solution of the present invention, in step 200, the translation model adopts an encoder-decoder structure, wherein the encoder encodes the input professional vocabulary into an intermediate representation, and the decoder decodes the intermediate representation into easy-to-understand text. The implementation method of using the database to train the translation model is as follows:

[0031] Selecting a model of encoder-decoder structure as a translation model, and dividing the vocabulary in the database into a training set and a validation set;

[0032] Convert each professional word and its popular explanation in the vocabulary of the database into an input format acceptable to the translation model, and calculate the difference between the output sequence of the translation model and the real sequence using a loss function;

[0033]

[0034] Where yi is the true popular explanation, pi is the accurate probability of the popular explanation generated by the translation model, and N represents the number of samples;

[0035] Based on the calculation result of the loss function, parameters of the encoder and decoder of the translation model are fine-tuned respectively until the loss value of the output sequence of the translation model meets expectations.

[0036] As a preferred solution of the present invention, a specific implementation method of fine-tuning the encoder of the translation model based on the calculation result of the loss function is:

[0037] Inputting the professional vocabulary in the vocabulary into the encoder of the translation model;

[0038] The encoder encodes the professional vocabulary into an intermediate representation Z;

[0039] Compare the intermediate representation Z generated by the encoder with the desired intermediate representation Z true Compare and calculate the loss value Loss dencoder =||ZZ true || 2 ;

[0040] Based on the calculated loss value, updating the parameters of the encoder,

[0041] As a preferred solution of the present invention, a specific implementation method of fine-tuning the decoder of the translation model based on the calculation result of the loss function is:

[0042] Inputting the intermediate representation generated by the encoder into the decoder;

[0043] The decoder gradually generates a sequence of words in the target language until a complete popular interpretation is generated;

[0044] Compare the generated popular interpretation y with the expected popular interpretation y t Compare and calculate the loss value Where yt is the true popular explanation, pt is the accurate probability of the popular explanation generated by the translation model, and T represents the number of samples;

[0045] Based on the calculated loss value, updating the parameters of the decoder,

[0046] As a preferred solution of the present invention, in step 300, the knowledge base is constructed, the interpretation translation content and the document content are vectorized, and the most relevant documents corresponding to the keywords in the professional vocabulary interpretation translation content are retrieved from the knowledge base using a vector similarity matching algorithm. The specific implementation method is:

[0047] Data entry: Enter the collected data into the knowledge base in a unified format. Each collected data includes document title, author, abstract, keywords, and full text link;

[0048] Data processing: Use natural language processing technology and machine learning algorithms to segment documents, extract key information and summarize, and classify the collected data, add keyword tags corresponding to each data, and use keyword tags to index each document in the knowledge base;

[0049] Data storage: using word embedding technology to convert text data obtained from data processing into vector form and store it in the knowledge base;

[0050] Matching query: Use the TF-IDF method to vectorize the documents, proofread them according to different similarities, and retrieve the most relevant documents corresponding to the keywords in the professional vocabulary interpretation translation content from the knowledge base.

[0051] As a preferred solution of the present invention, in step 400, after retrieving the document most relevant to the user's query from the knowledge base, the implementation steps of extracting key information in the most relevant document are:

[0052] Document preprocessing: decompose the text into words or phrases, convert them into a unified text format, remove irrelevant format information, and mark the parts of speech of the decomposed single words or short sentences;

[0053] Information identification and extraction: Use TF-IDF or TextRank to extract the most relevant keywords in the document. NER is used to identify technical parameters and equipment name entities, and the relationship between entities is analyzed to extract data values ​​and chart information from the experimental results.

[0054] Summarization: Generate document summaries using extractive or generative summarization methods;

[0055] Information structuring: Organizing the extracted information into a structured form;

[0056] Information extraction: Use entity recognition and relationship extraction techniques to extract key information.

[0057] In addition, the present invention also provides a translation and analysis system for oil and gas geological survey professional vocabulary translation and analysis method, including:

[0058] User inquiry operation interface: used to input user questions, supporting text and voice input;

[0059] A professional vocabulary extraction unit, used to extract professional vocabulary from questions asked by users;

[0060] Professional vocabulary translation unit, used to translate professional vocabulary into easy-to-understand content;

[0061] A knowledge base matching unit extracts keywords based on the content of the professional vocabulary translation unit, vectorizes the keywords and documents of the knowledge base content, and retrieves the most relevant documents to the user query from the knowledge base by means of vector similarity matching;

[0062] Key information extraction and generation module: extracts key information from documents most relevant to user queries and generates coherent answers;

[0063] User interface and feedback module: provides user interaction interface and feedback channels.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] (1) Improving the ability to identify and translate professional terms: By designing a professional vocabulary recognition algorithm and translation module (a fine-tuned oil and gas geological survey model), the system can accurately identify and translate professional terms in the field of oil and gas geology, converting complex professional terms into easy-to-understand language so that non-professionals can also understand the core content of the information, thereby improving the comprehensibility and accuracy of the information and promoting effective communication and collaboration between different disciplines.

[0066] (2) Optimizing the knowledge base matching process: Through keyword fine-tuning and professional vocabulary translation, the system can more accurately match relevant documents from the knowledge base, form concise and clear answers, reduce false matches and missed matches, ensure that users obtain the most relevant and accurate information, and improve user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0068] Figure 1 A flowchart of a professional intelligent question-answering method according to an embodiment of the present invention; DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] like Figure 1 As shown, the present invention provides a professional intelligent question-answering method for oil and gas geological survey, comprising the following steps:

[0071] Step 100: Receive a question input by a user, and extract professional vocabulary from the input question.

[0072] In step 100, the method for extracting professional vocabulary from the input question is:

[0073] Step 101, collect relevant documents of oil and gas geological survey to establish a professional vocabulary database;

[0074] Step 102: Capture the context information and semantic features of the question input by the user, and identify the professional vocabulary in the question input by the user in combination with a predefined rule base.

[0075] In step 102, the user inputs questions through the user inquiry operation interface, which supports text and voice input. The user inquiry operation interface transmits the user input signal to the processor (Chat model), and the Chat model calls the keyword extraction unit to extract professional vocabulary from the questions input by the user.

[0076] In order to extract professional vocabulary from the questions input by the user, the implementation method of establishing a professional vocabulary library is as follows:

[0077] A large amount of document materials related to oil and gas geological surveys are collected, including but not limited to literature libraries, research reports, academic papers, etc. By analyzing and processing these documents, a professional vocabulary library containing specific terms and terminology in the field of oil and gas geology is constructed.

[0078] In order to further realize the next step of translation and interpretation of each professional vocabulary, in this step, the corresponding interpretation and translation content of each professional vocabulary is integrated into the professional vocabulary library, and the professional vocabulary in the professional vocabulary library is marked, so that the professional vocabulary with consistent marking is retained in the professional vocabulary library, and the professional vocabulary with inconsistent marking is removed.

[0079] After integrating the translated explanations corresponding to each professional term into the professional term library, a glossary and a relationship table are formed. Among them, the glossary stores each professional term and the corresponding translated explanations, and the relationship table stores the association relationships (including synonyms, antonyms, and hyponymy relationships) between different professional terms.

[0080] After constructing the professional term library, invite domain experts to annotate the professional terms in the library to ensure the accuracy and consistency of the annotation. Secondly, make a professional judgment on the professional terms in the professional term library, including calculating the word frequency of each professional term, part-of-speech tagging, and extracting context information.

[0081] When capturing and identifying professional terms in the user's input question in combination with the professional term library, the cleaning work is a very important step, which ensures the accuracy and efficiency of subsequent processing. The specific implementation method of the cleaning work is as follows:

[0082] ① Word segmentation and sentence segmentation: Perform word segmentation on the user's input to break it down into individual words or phrases. This step is particularly important for Chinese because Chinese has no obvious word boundaries. Commonly used Chinese word segmentation tools include jieba, HanLP, etc.

[0083] ② Stop word removal: Establish a stop word list to filter out these words. Stop words refer to those words that frequently appear in the text but are not very helpful for understanding the text content, such as "de", "shi", "zai", etc. These words usually will not be retained as keywords.

[0084] ③ Part-of-speech tagging: Perform part-of-speech tagging on the results of word segmentation, which helps to better understand the role of each word in the sentence. Commonly used part-of-speech tagging tools include Stanford POS Tagger, NLTK, spaCy, etc.

[0085] ④ Matching with the professional term library: Match the results of word segmentation and part-of-speech tagging with the pre-established professional term library. Only those words that exist in the professional term library will be retained. This step ensures that the retained words are all professional terms related to oil and gas geological survey.

[0086] ⑤ Context analysis: Even if some words exist in the professional term library, it is necessary to further consider their meanings in the specific context. For example, the word "pressure" may have different meanings in different situations. Through context analysis, it is possible to more accurately judge whether a certain word should be retained.

[0087] ⑥ Applying the rule library: Combine the pre-defined rule library to further screen the remaining words. For example, rules can be set to exclude those words that, although belonging to professional terms, are irrelevant to the current question.

[0088] The example process is as follows:

[0089] Assume the user input question is: “How to evaluate the reservoir characteristics of an oil and gas field?”;

[0090] Word segmentation: How to evaluate the reservoir characteristics of an oil and gas field;

[0091] Remove stop words: evaluation / oil and gas field / reservoir / characteristics;

[0092] Part-of-speech tagging: verb / noun / noun / noun;

[0093] Professional vocabulary database matching: Assuming that "assessment", "oil and gas field", "reservoir", and "characteristics" are all in the professional vocabulary database, all of them are retained.

[0094] Context analysis: Confirm that “assessment,” “field,” “reservoir,” and “properties” all make sense in the context of the problem at hand.

[0095] Apply the rule base validation: Assuming that there are no rules in the rule base to further exclude these words, the final retained words are "assessment", "oil and gas field", "reservoir", and "characteristics".

[0096] Through the above steps, keywords can be effectively extracted from the questions input by users, ensuring the accuracy and efficiency of subsequent processing.

[0097] When identifying professional vocabulary in user input questions, it is necessary to select a suitable Chat model to learn the data, such as using a rule matching method to identify professional vocabulary based on a predefined rule base. Use supervised (mechanical) learning methods, such as support vector machines (SVM) and random forests, to train a classifier to identify professional vocabulary, and use deep learning models, such as Bi-LSTM and Transformer, to capture the contextual information and semantic features of vocabulary and improve recognition accuracy.

[0098] Furthermore, based on the pre-trained general language model, the extracted professional vocabulary is fine-tuned to form professional vocabulary that is the same as the vocabulary in the professional vocabulary library. The general language model is trained with a large amount of professional literature and annotated data to ensure the accuracy of the fine-tuned professional vocabulary in specific fields.

[0099] Specifically select a pre-trained language model (such as BERT, RoBERTa, T5, etc.). These models have been pre-trained on large-scale corpora and have strong language understanding and generation capabilities. Based on the pre-trained language model, use professional data in the field of oil and gas geological surveys to fine-tune the keywords extracted from user input questions to turn them into professional vocabulary.

[0100] After selecting the Chat model that recognizes professional vocabulary in user input questions, preprocess the user input questions, including word segmentation, stop word removal, stemming, and other operations to ensure the cleanliness and consistency of the data. Extract the frequency, part of speech, context information and other features of the professional vocabulary in the user input questions, build feature vectors, use the user input questions to train the model, optimize the model parameters, and improve the recognition accuracy.

[0101] In order to verify the factors affecting the accuracy of the responses of the entire question-answering system by using the Chat model to identify professional vocabulary in user input questions, a test set of 1,000 oil and gas geological survey documents was used to evaluate the system without and with professional vocabulary recognition algorithms. The experimental results show that by using the professional vocabulary recognition algorithm, the system's performance in professional vocabulary recognition is significantly improved, with the accuracy and recall rates increased by 20% and 20% respectively, and the F1 score increased by 20%.

[0102] Without professional vocabulary recognition algorithm: the accuracy of professional vocabulary recognition is 70%, the recall rate is 65%, and the F1 score is 0.67. With professional vocabulary recognition algorithm: the accuracy of professional vocabulary recognition is 90%, the recall rate is 85%, and the F1 score is 0.87.

[0103] Step 200: Using a translation model to interpret and translate the extracted professional vocabulary to form a popular explanation of the professional vocabulary.

[0104] Regarding the explanation of professional terms, relevant professional data are collected from user feedback, expert knowledge, academic papers, textbooks, technical reports, journal reports, etc. These data contain a professional term and its corresponding popular explanation. For example, the popular explanation of "reservoir" is "the stratum that stores oil and gas"; the corresponding explanation of "geological structure" is "the structure and morphology inside the earth's crust."

[0105] In step 200, the translation model adopts an encoder-decoder structure, wherein the encoder encodes the input professional vocabulary into an intermediate representation, and the decoder decodes the intermediate representation into easy-to-understand text. The implementation method of using the database to train the translation model is:

[0106] Selecting a model of encoder-decoder structure as a translation model, and dividing the vocabulary in the database into a training set and a validation set;

[0107] Convert each professional word and its popular explanation in the vocabulary of the database into an input format acceptable to the translation model, and calculate the difference between the output sequence of the translation model and the real sequence using a loss function;

[0108]

[0109] Where yi is the true popular explanation, pi is the accurate probability of the popular explanation generated by the translation model, and N represents the number of samples;

[0110] Based on the calculation result of the loss function, parameters of the encoder and decoder of the translation model are fine-tuned respectively until the loss value of the output sequence of the translation model meets expectations.

[0111] The specific implementation method of fine-tuning the encoder of the translation model based on the calculation result of the loss function is:

[0112] Inputting the professional vocabulary in the vocabulary into the encoder of the translation model;

[0113] The encoder encodes the professional vocabulary into an intermediate representation Z;

[0114] Compare the intermediate representation Z generated by the encoder with the desired intermediate representation Z true Compare and calculate the loss value Loss dencoder =||ZZ true || 2 ;

[0115] Based on the calculated loss value, updating the parameters of the encoder,

[0116] The specific implementation method of fine-tuning the decoder of the translation model based on the calculation result of the loss function is:

[0117] Inputting the intermediate representation generated by the encoder into the decoder;

[0118] The decoder gradually generates a sequence of words in the target language until a complete popular interpretation is generated;

[0119] Compare the generated popular interpretation y with the expected popular interpretation y t Compare and calculate the loss value Where yt is the true popular explanation, pt is the accurate probability of the popular explanation generated by the translation model, and T represents the number of samples;

[0120] Based on the calculated loss value, updating the parameters of the decoder,

[0121] Based on the above, during the experimental operation, after translating professional vocabulary in a popular and understandable way, the matching key content of the document was retrieved from various documents based on the translated vocabulary, which can improve the accuracy of professional vocabulary translation. The experiment used a test set containing 1,000 professional vocabulary for oil and gas geological surveys, and evaluated the system without using the professional vocabulary interpretation model and the system using the professional vocabulary interpretation model. Experimental results: By using the professional vocabulary interpretation model, the performance of the system in professional vocabulary interpretation was significantly improved, the accuracy was increased by 25%, and the BLEU score and ROUGE score were increased by 20% and 20% respectively.

[0122] Without the professional vocabulary explanation model: the accuracy of the professional vocabulary explanation is 60%, the BLEU score is 0.65, and the ROUGE score is 0.60. With the professional vocabulary explanation model: the accuracy of the professional vocabulary explanation is 85%, the BLEU score is 0.85, and the ROUGE score is 0.80.

[0123] Through step 200, this embodiment can improve the ability to recognize and translate professional terms. By designing a professional vocabulary recognition algorithm and a translation module (a fine-tuned oil and gas geological survey model), the system can accurately recognize and translate professional terms in the field of oil and gas geology, and convert complex professional terms into easy-to-understand language, so that non-professionals can also understand the core content of the information, thereby improving the comprehensibility and accuracy of the information and promoting effective communication and collaboration between different disciplines.

[0124] Step 300: construct a knowledge base in the field of oil and gas geological survey, extract keywords from popular explanations of professional terms, and use the keywords as search conditions to match relevant documents from the knowledge base.

[0125] In step 300, after the professional vocabulary is interpreted into easy-to-understand text, paragraph-style text is formed. In order to achieve matching retrieval of the corresponding parsed document from the knowledge base, it is necessary to screen out the keywords for matching retrieval from the interpreted easy-to-understand text, thereby ensuring the relevance, representativeness and uniqueness of the keywords and improving the accuracy and coverage of the retrieval.

[0126] The screening principles include the following:

[0127] Relevance: Keywords should be closely related to the core meaning of professional vocabulary and accurately reflect the meaning of professional vocabulary. For example, the keywords extracted from "reservoir" include "reservoir", "oil and gas", "stratum", etc. These keywords directly reflect the definition of reservoir.

[0128] Representativeness: Keywords should be highly representative and able to summarize the main content of the paragraph. For example, the keywords extracted from "lithology" include "lithology", "rock", "property", "type", and "mineral", which cover the main characteristics of lithology.

[0129] Uniqueness: Keywords should have a certain degree of uniqueness and avoid overly common words in order to improve the accuracy of the search. The keywords extracted from "geological structure" include "geological structure", "crust", "structure", "morphology", and "surface". These keywords have high uniqueness and avoid common words such as "is" and "of".

[0130] Frequency: Keywords appear more frequently in a paragraph, which helps improve the coverage of the search. For example, the keywords extracted from "reservoir", "oil and gas", "stratum", "porosity", and "rock" appear multiple times in the paragraph, ensuring the coverage of the search.

[0131] Based on the above, the specific implementation steps and methods for selecting keywords from the explanation sentences of professional vocabulary are as follows:

[0132] (1) Preprocessing: Perform preprocessing operations such as word segmentation, stop word removal, and stem extraction on paragraphs to ensure data consistency and reduce inventory.

[0133] (2) Extract keywords;

[0134] Method 1: TF-IDF (Term Frequency-Inverse Document Frequency): A statistical method used to evaluate the importance of a word in a document or corpus.

[0135] Steps: Calculate the term frequency (TF) of each word in the paragraph; Calculate the inverse document frequency (IDF) of each word in the entire knowledge base; Calculate the TF-IDF value of each word; Select several words with the highest TF-IDF values ​​as keywords.

[0136] Method 2: TextRank: A graph-based ranking algorithm for extracting keywords from text.

[0137] Steps: Construct a word co-occurrence graph, where the nodes are words and the edge weights are the co-occurrence frequencies between words; use the PageRank algorithm to calculate the weight of each node; select several words with the highest weights as keywords.

[0138] Method 3: Rule-based method: Manually filter keywords according to pre-defined rules.

[0139] Steps: Define the part of speech of the keyword (such as noun, verb, adjective, etc.); use natural language processing tools (such as NLTK, spaCy) to extract the part of speech tags in the paragraph; select words that meet the rules as keywords.

[0140] Method 4: Model-based approach: Automatically extract keywords using machine learning or deep learning models.

[0141] Steps: Prepare a labeled dataset containing paragraphs and corresponding keywords; train a classification model (such as SVM, BERT, etc.) to classify each word as a keyword or non-keyword; use the trained model to extract keywords from new paragraphs.

[0142] Furthermore, in step 300, a comprehensive, accurate and easy-to-search professional vocabulary database in the field of oil and gas geological survey is constructed to provide efficient information support for researchers and engineers.

[0143] The process of building a knowledge base:

[0144] Determine the scope of the knowledge base: Clarify the subject areas covered by the knowledge base, such as exploration technology, geological modeling, reservoir management, etc.

[0145] Collect information: Collect relevant information from academic journals, conference papers, industry reports, technical manuals and other sources.

[0146] Data entry: Enter the collected information into the knowledge base in a unified format, including document title, author, abstract, keywords, full-text link and other information.

[0147] Data processing: Create an index for each document in the knowledge base to ensure quick retrieval by keyword. Categorize documents and add appropriate tags to facilitate users to find materials based on different topics or types. Use natural language processing technology and machine learning algorithms to segment documents, extract key information and summarize, and generate structured data of question + response type;

[0148] Review and proofreading: Experts in the field review the entered documents to ensure the accuracy and reliability of the information.

[0149] Data storage: Use word embedding technology to convert text data into vector form and store it in a vector database.

[0150] Continuous updating: Regularly update the knowledge base, add new research results and industry trends, and ensure the timeliness and completeness of the knowledge base.

[0151] User feedback: Collect user feedback on the knowledge base and continuously optimize the system's functionality and user experience.

[0152] Matching algorithm: Use vector similarity matching algorithm to retrieve the most relevant documents from the knowledge base. Vocabulary comparison table: A dynamically updated vocabulary comparison table or database that stores the relationship between professional vocabulary and its corresponding explanation.

[0153] The experiment shows that by using the vocabulary translation database, the system's performance in vocabulary translation is significantly improved, with the accuracy increased by 20%, and the recall and precision rates increased by 20% and 15% respectively.

[0154] This implementation method can more accurately match relevant documents from the knowledge base through keyword fine-tuning and professional vocabulary translation, form concise and clear answers, reduce false matches and missed matches, ensure that users obtain the most relevant and accurate information, and improve user satisfaction.

[0155] Accurate terminology translation reduces decision errors caused by misunderstanding of terminology, avoiding unnecessary rework and increased costs. By automatically extracting and summarizing non-standardized documents, the integrity and accuracy of the knowledge base is ensured, thereby improving the reliability of decisions made based on these data.

[0156] Step 400: extract key information from relevant documents, and feed back the extracted key information to the user through the front-end interface.

[0157] (1) Extraction of key information: Use natural language technology (publicly known NLP technology) to extract key information from relevant documents, including technical parameters, experimental results, and experimental conclusions. NLP technology can extract the core ideas or conclusions, abstracts, keywords, themes, and other information of the document. Implementation steps:

[0158] Document preprocessing: decompose the text into words or phrases, convert them into a unified text format, remove irrelevant format information, and annotate the parts of speech, such as "assessment / oil and gas field / reservoir", part of speech tagging: verb / noun / noun.

[0159] Information identification and extraction: Use TF-IDF or TextRank to extract keywords. Apply NER to identify entities such as technical parameters and equipment names. Analyze the relationship between entities. Extract data values ​​and chart information from experimental results.

[0160] Summary generation: Generate a document summary using extractive or generative summarization methods. Ensure that the summary contains key technical parameters, experimental results, and conclusions.

[0161] Information structuring: Organize the extracted information into a structured form, such as tables, lists, etc.

[0162] Front-end interface display: Design a user-friendly front-end interface. Highlight key information and provide options for viewing detailed content.

[0163] (2) Vector similarity matching: Use the TF-IDF method to vectorize the documents, proofread them according to different similarities, and retrieve the documents most relevant to the user query from the knowledge base.

[0164] (3) Information extraction: Use entity recognition and relationship extraction techniques to extract key information.

[0165] Named Entity Recognition (NER) and Relation Extraction (RE) technologies are public and well-known technologies that have been widely studied and applied in academia and industry, and there are many open source tools and frameworks available, such as spaCy, Stanford NLP, BERT, etc. Its working principle: identify entities based on predefined rules and patterns; use machine learning algorithms (such as CRF, HMM, SVM, etc.) to train models and learn entity features through labeled data sets; use neural networks (such as BiLSTM-CRF, BERT, etc.) for entity recognition and relation extraction, and these models can capture more complex contextual information.

[0166] Implementation steps: First, perform preprocessing operations such as word segmentation, stop word removal, stem extraction, and lemmatization on the document; use the NER model and RE model to identify entities in the document (technical parameters, equipment names, etc.) and the relationships between entities (causal relationships, comparative relationships, etc.); then organize the extracted information into a structured form, such as tables, lists, etc.; finally, display the information to the user through the front-end interface.

[0167] (4) Generate answers: Generate coherent answers based on the extracted key information, ensuring the accuracy and readability of the answers. Use natural language generation technology to generate user-friendly answers.

[0168] (5) User interface feedback: The generated answers are fed back to the user through the front-end interface. Provide user feedback channels, collect user opinions, and continuously optimize system performance.

[0169] Among them, NLP technology and entity recognition and relationship extraction technology are two different methods, but they can be used complementary to ensure that the extracted information is both comprehensive and accurate. The key information extracted by these two technologies may overlap, but they may also cover different aspects: NLP technology is mainly used to understand and process natural language texts, and can extract more generalized information such as the main points, summaries, and sentiment analysis in the text. Entity recognition and relationship extraction technology focus on identifying specific entities in the text and the relationships between these entities.

[0170] Their complementarity is reflected in the fact that they may both extract some of the same keywords or entities, but NLP technology focuses more on overall understanding and summary generation, providing an overall understanding of the text; while entity recognition and relationship extraction technology focus more on the identification of specific details and relationships, and can extract specific technical parameters and experimental results mentioned in the summary.

[0171] In the user satisfaction survey, the user satisfaction of the system of the present invention increased by 25%. The frequency of use of users increased significantly. During the one-month observation period, the average daily use of the system of the present invention was 150 times, while the average daily use of the prior art was 100 times.

[0172] The translation system is integrated with the knowledge base to quickly and accurately parse professional terminology, reducing the time spent on understanding and interpreting terminology, allowing the team to reach consensus and advance projects more quickly, raising the understanding threshold for non-professionals and the user-friendliness of the system.

[0173] In addition, the present invention also provides a translation and analysis system for oil and gas geological survey professional vocabulary translation and analysis method, including:

[0174] User inquiry operation interface: used to input user questions, supporting text and voice input;

[0175] A professional vocabulary extraction unit, used to extract professional vocabulary from questions asked by users;

[0176] Professional vocabulary translation unit, used to translate professional vocabulary into easy-to-understand content;

[0177] A knowledge base matching unit extracts keywords based on the content of the professional vocabulary translation unit, vectorizes the keywords and documents of the knowledge base content, and retrieves the most relevant documents to the user query from the knowledge base by means of vector similarity matching;

[0178] Key information extraction and generation module: extracts key information from documents most relevant to user queries and generates coherent answers;

[0179] User interface and feedback module: provides user interaction interface and feedback channels.

[0180] This implementation provides an efficient and coherent translation process, including steps such as user input processing, professional vocabulary recognition, vocabulary translation, document matching and result generation. When a user asks a question, the system can automatically identify the professional vocabulary in the question and expand the matching rate and matching accuracy of the output matching answer content by interpreting the professional vocabulary, providing a seamless user experience so that both professional and non-professional users can obtain accurate information.

[0181] Therefore, this embodiment can achieve the following technical effects:

[0182] (1) Improving the ability to identify and translate professional terms: By designing a professional vocabulary recognition algorithm and translation module (a fine-tuned oil and gas geological survey model), the system can accurately identify and translate professional terms in the field of oil and gas geology, converting complex professional terms into easy-to-understand language so that non-professionals can also understand the core content of the information, thereby improving the comprehensibility and accuracy of the information and promoting effective communication and collaboration between different disciplines.

[0183] (2) Optimizing the knowledge base matching process: Through keyword fine-tuning and professional vocabulary translation, the system can more accurately match relevant documents from the knowledge base, form concise and clear answers, reduce false matches and missed matches, ensure that users obtain the most relevant and accurate information, and improve user satisfaction.

[0184] (3) Reduce error rates and improve work efficiency: Accurate terminology translation reduces decision errors caused by misunderstanding of terminology, avoiding unnecessary rework and increased costs. By automatically extracting and summarizing non-standardized documents, the integrity and accuracy of the knowledge base is ensured, thereby improving the reliability of decisions made based on these data.

[0185] (4) Improve user experience: The translation system is integrated with the knowledge base to quickly and accurately parse professional terms, reducing the time spent on understanding and interpreting terms. This allows the team to reach consensus and advance the project more quickly, raising the understanding threshold for non-professionals and improving the user-friendliness of the system.

[0186] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A professional intelligent question-answering method for oil and gas geological survey, characterized in that: The following steps are involved: Step 100: receiving a question input by a user, and extracting professional vocabulary from the input question; Step 200: using a translation model to interpret and translate the extracted professional vocabulary to form a popular interpretation of the professional vocabulary; Step 300: construct a knowledge base in the field of oil and gas geological survey, extract keywords from popular explanations of professional terms, and use the keywords as search conditions to match relevant documents from the knowledge base; Step 400: extract key information from relevant documents, and feed back the extracted key information to the user through the front-end interface.

2. A professional intelligent question-answering method for oil and gas geological survey according to claim 1, characterized in that: In step 100, the method for extracting professional vocabulary from the input question is: Collect relevant documents of oil and gas geological survey to establish a professional vocabulary database; Capture the contextual information and semantic features of the questions entered by the user, and combine with the customized rule base to identify the professional vocabulary in the questions entered by the user.

3. A professional intelligent question-answering method for oil and gas geological survey according to claim 2, characterized in that: The steps to build a professional vocabulary database are: Collect a large amount of documents related to oil and gas geological surveys, analyze and process the documents, and build a professional vocabulary database containing special terms and terminology in the field of oil and gas geology; The explanation and translation content corresponding to each professional vocabulary is integrated into the professional vocabulary library, and the professional vocabulary in the professional vocabulary library is annotated, so that the professional vocabulary with consistent annotations is retained in the professional vocabulary library, and the professional vocabulary with inconsistent annotations is removed.

4. A professional intelligent question-answering method for oil and gas geological survey according to claim 2 or 3, characterized in that: The steps for capturing and identifying professional vocabulary in the questions input by the user in combination with the professional vocabulary library are as follows: Perform word segmentation on the questions input by the user, breaking the questions input by the user into multiple independent words or phrases; Create a stop word list to filter out stop words in questions entered by users; Perform part-of-speech tagging on the results after word segmentation to distinguish the decomposed words into nouns, verbs or adjectives; Matching the words after word segmentation and part-of-speech tagging with a pre-established professional vocabulary database, and extracting the same professional vocabulary as that in the professional vocabulary database from the words after word segmentation and part-of-speech tagging; Combined with context analysis, two adjacent professional vocabulary are further combined to form an overall vocabulary, and the overall vocabulary is matched with a pre-established professional vocabulary library; The custom rule base further filters the extracted professional vocabulary that is the same as that in the professional vocabulary base to exclude the vocabulary that belongs to the professional vocabulary but is irrelevant to the current problem.

5. A professional intelligent question-answering method for oil and gas geological survey according to claim 1, characterized in that: In step 200, the translation model adopts an encoder-decoder structure, wherein the encoder encodes the input professional vocabulary into an intermediate representation, and the decoder decodes the intermediate representation into easy-to-understand text. The implementation method of using the database to train the translation model is: Selecting a model of encoder-decoder structure as a translation model, and dividing the vocabulary in the database into a training set and a validation set; Convert each professional word and its popular explanation in the vocabulary of the database into an input format acceptable to the translation model, and calculate the difference between the output sequence of the translation model and the real sequence using a loss function; Where yi is the true popular explanation, pi is the accurate probability of the popular explanation generated by the translation model, and N represents the number of samples; Based on the calculation result of the loss function, parameters of the encoder and decoder of the translation model are fine-tuned respectively until the loss value of the output sequence of the translation model meets expectations.

6. A professional intelligent question-answering method for oil and gas geological survey according to claim 5, characterized in that: The specific implementation method of fine-tuning the encoder of the translation model based on the calculation result of the loss function is: Inputting the professional vocabulary in the vocabulary into the encoder of the translation model; The encoder encodes the professional vocabulary into an intermediate representation Z; Compare the intermediate representation Z generated by the encoder with the desired intermediate representation Z true Compare and calculate the loss value Loss dencoder =||ZZ true || 2 ; Based on the calculated loss value, update the encoder parameters, θ dencoder * = θ dencoder - 7. A professional intelligent question-answering method for oil and gas geological survey according to claim 5, characterized in that: The specific implementation method of fine-tuning the decoder of the translation model based on the calculation result of the loss function is: Inputting the intermediate representation generated by the encoder into the decoder; The decoder gradually generates a sequence of words in the target language until a complete popular interpretation is generated; Compare the generated popular interpretation y with the expected popular interpretation y t Compare and calculate the loss value Where yt is the true popular explanation, pt is the accurate probability of the popular explanation generated by the translation model, and T represents the number of samples; Based on the calculated loss value, updating the parameters of the decoder, 8. A professional intelligent question-answering method for oil and gas geological survey according to claim 5, characterized in that: In step 300, the knowledge base is constructed, the interpretation translation content and the document content are vectorized, and the most relevant documents corresponding to the keywords in the professional vocabulary interpretation translation content are retrieved from the knowledge base using a vector similarity matching algorithm. The specific implementation method is: Data entry: Enter the collected data into the knowledge base in a unified format. Each collected data includes document title, author, abstract, keywords, and full text link; Data processing: Use natural language processing technology and machine learning algorithms to segment documents, extract key information and summarize, and classify the collected data, add keyword tags corresponding to each data, and use keyword tags to index each document in the knowledge base; Data storage: using word embedding technology to convert text data obtained from data processing into vector form and store it in the knowledge base; Matching query: Use the TF-IDF method to vectorize the documents, proofread them according to different similarities, and retrieve the most relevant documents corresponding to the keywords in the professional vocabulary interpretation translation content from the knowledge base.

9. A professional intelligent question-answering method for oil and gas geological survey according to claim 8, characterized in that: In step 400, after retrieving the document most relevant to the user's query from the knowledge base, the steps for extracting key information in the most relevant document are: Document preprocessing: decompose the text into words or phrases, convert them into a unified text format, remove irrelevant format information, and mark the parts of speech of the decomposed single words or short sentences; Information identification and extraction: Use TF-IDF or TextRank to extract the most relevant keywords in the document. NER is used to identify technical parameters and equipment name entities, and the relationship between entities is analyzed to extract data values ​​and chart information from the experimental results. Summarization: Generate document summaries using extractive or generative summarization methods; Information structuring: Organizing the extracted information into a structured form; Information extraction: Use entity recognition and relationship extraction techniques to extract key information.

10. A translation and analysis system based on the oil and gas geological survey professional vocabulary translation and analysis method according to any one of claims 1 to 9, characterized in that: include: User inquiry operation interface: used to input user questions, supporting text and voice input; A professional vocabulary extraction unit, used to extract professional vocabulary from questions asked by users; Professional vocabulary translation unit, used to translate professional vocabulary into easy-to-understand content; A knowledge base matching unit extracts keywords based on the content of the professional vocabulary translation unit, vectorizes the keywords and documents of the knowledge base content, and retrieves the most relevant documents to the user query from the knowledge base by means of vector similarity matching; Key information extraction and generation module: extracts key information from documents most relevant to user queries and generates coherent answers; User interface and feedback module: provides user interaction interface and feedback channels.

Images