274 results about "Natural language generation" patented technology

The invention provides a method for generating an SQL (Structured Query Language) by a natural language based on bidirectional mapping and semantic parsing, which relates to the technical field of database query and comprises the following steps of: extracting natural language query elements and packaging the natural language query elements into structured data, and establishing a mapping relationship from a query field to a service attribute and a physical data table by adopting a bidirectional Hash index technology; and automatically identifying multi-table association keys, performing semantic extension and compliance verification, generating an abstract syntax tree, performing processing according to user permission, and finally converting the abstract syntax tree into an SQL statement conforming to a target database syntax specification. According to the method, the accuracy and the efficiency of converting the natural language into the SQL are improved, and the flexibility and the safety of the system are enhanced.

The invention specifically relates to the technical field of natural language processing, and discloses an intelligent interactive customer service system based on an AI large model, which comprises a multi-modal input analysis module, an intention preliminary detection module, a large model semantic understanding module, a dynamic knowledge base association module, a dialogue strategy generation module, a natural language generation module and a service feedback work order module, through deep cooperation of a multi-modal input analysis module and a large model semantic understanding module, entity and attribute information is accurately extracted, context semantic association information is mined, and through the cooperative effect of an intention preliminary detection module, a dynamic knowledge base association module and a dialogue strategy generation module, based on a semantic analysis result output by a large model, a dialogue strategy is generated. According to the method and the system, matching calculation is carried out with the constructed dynamic knowledge base, rich knowledge support is provided for solving complex problems, personalized knowledge content is recommended for different users through the natural language generation module and the service feedback work order module, and the reply matching degree and the complex problem processing capacity are greatly improved.

The invention discloses a consultation method and system based on natural language processing and a legal knowledge graph, and relates to the field of data processing, and the method comprises the steps: receiving a multi-format legal consultation demand of a user, converting the multi-format legal consultation demand into a text, inputting the text into a BERT law NLP model, and analyzing key information through word segmentation, intention recognition and entity extraction; based on a pre-constructed multi-level legal knowledge graph, carrying out accurate and fuzzy retrieval and domain filtering in combination with an analysis result, and obtaining an associated law article, a case and a legal relationship; screening conflict law articles and similar cases, and inputting the conflict law articles and the similar cases into a graph neural network reasoning model to generate a preliminary conclusion; the conclusion is converted into a spoken consultation report through a natural language generation module, and output is customized according to a user scene; and if the user feedback satisfaction degree is less than the threshold value, iteratively optimizing the storage data to the historical library. The method has the advantages that accurate retrieval is realized based on the BERT model and the multi-level knowledge graph in the legal field, the oral personalized conclusion combined with the user scene is generated through GNN reasoning, and iterative optimization is performed through user feedback.

The invention discloses an electrical work ticket intelligent generation method and device, a medium and equipment in the technical field of power systems, and aims to solve the problems of serious module decoupling, dependence on manual connection and insufficient intelligent generation capability in the prior art. The method comprises the following steps: generating a required digital knowledge base and an electrical work task description generation work ticket based on a pre-constructed work ticket by utilizing a multi-model collaborative generation architecture; the multi-model collaborative generation architecture comprises a rule driving module used for calling a corresponding template based on task element matching; the task execution unit is used for executing a specific subtask according to a pre-deployed lightweight model or rule engine; and the large language model is used for performing semantic understanding and reasoning according to the electrical work task description, calling the rule driving module and the task execution unit according to a semantic understanding and reasoning result, and generating a work ticket in combination with a digital knowledge base required by work ticket generation. The method has both the accuracy of structured knowledge recognition and the flexibility of natural language generation.

Techniques for hybrid natural language generation (NLG) using a symbolic NLG engine and a large language model (LLM) are disclosed. The techniques include: obtaining a first text segment configuration; obtaining values of at least some of a first set of data variables; generating a first natural language text segment for an electronic document using a first text segment configuration, the values, and the symbolic NLG engine; obtaining an initial text segment; obtaining a second text segment configuration specifying information to use for generating a second natural language text segment using the ML NLG engine from the initial text segment; generating a second natural language text segment for the electronic document by using the initial text segment, the second text segment configuration, and the ML NLG engine; and outputting the generated electronic document.

The invention discloses a virtual power plant user incentive generation method and system based on a deep user portrait, and relates to the technical field of power demand side management. The method comprises the steps of collecting multi-source heterogeneous data and performing fusion preprocessing to form a multi-source user data set; generating a deep user portrait according to the multi-source user data set, and generating a personalized incentive strategy by using a reinforcement learning algorithm in combination with a power grid operation state and a historical response record; determining an optimal push time period through a user behavior activity prediction model, generating an incentive push message by using a natural language generation method, and actively pushing the incentive push message to a user terminal; and updating the user portrait and the incentive strategy according to the user response result and the feedback information to form closed-loop optimization. According to the method, accurate description and differential excitation of user side response behaviors are realized, privacy compliance and cost effectiveness are considered, and the adjustable capacity and scheduling reliability of the virtual power plant are enhanced while the participation rate and the response quality are improved.

A method includes preparing a base model using an input model pretrained on at least three languages different from each other and a base vocabulary including words corresponding to two languages among the at least three languages, where the preparing the base model includes constraining the input model to the words included in the base vocabulary; training the base model using a first enhanced training dataset generated from public data, to generate a text summarization model; training the base model using a second enhanced training dataset generated from the first enhanced training dataset, to generate a text generation model; and training the base model using a third enhanced training dataset that is generated using the second enhanced training dataset and the text summarization model, to generate a next sentence generation model.

The invention belongs to the technical field of artificial intelligence and educational informatization, and relates to a multi-mode knowledge graph enhanced intelligent teaching-assistant question-answering system. According to the method, the knowledge graph construction technology, the multi-modal content analysis technology and the large language model reasoning enhancement technology are comprehensively applied, and the semantic understanding, knowledge integration and reasoning generation capabilities of the intelligent teaching assisting system in an education and teaching scene are improved. The related technology comprises layout analysis of textbook documents, semantic description generation of image content, entity and relation extraction of text content, multi-modal knowledge graph construction and question and answer reasoning and natural language generation combined with the knowledge graph. Through cooperative application of the technologies, semantic interconnection can be carried out on various modal information such as texts, images and tables in the textbook, and a searchable and traceable textbook-level knowledge network is formed.

The invention provides a mathematical question and answer process teaching system, which comprises a user interaction module, a question analysis and standardization module, a cognitive diagnosis engine, a generative teaching module, a mathematical knowledge graph database and a user portrait database, teaching content and guiding questions are presented to the user in a multi-mode form; the question analysis and standardization module is used for performing semantic understanding and structured analysis on mathematical questions and generating standardized question description and at least one standard question solving path; according to the method, the structured reasoning of the knowledge graph is combined with the natural language generation ability of the generative AI, so that a closed loop from'problem solving 'to'diagnosis' to'heuristic teaching 'is realized, thinking blockage points of students can be accurately positioned and personalized tutoring can be provided like human teachers, and the learning efficiency and the autonomous thinking ability are remarkably improved.

The invention discloses a sign language translation model, a sign language translation system and a sign language translation method based on full-modal alignment. The sign language translation method comprises the steps of extracting multi-modal features of hand, face and body postures from an input video and performing preliminary fusion; deep secondary fusion and alignment are carried out through multi-scale time sequence coding and a cross-modal collaborative attention mechanism, and a spatial-temporal feature sequence of full-modal alignment is generated; performing boundary detection and dynamic segmentation on the feature sequence by using a CTC-based sequence prediction model, and outputting a discrete sign language word sequence with a timestamp; and finally, capturing a sign language grammar structure of the sequence through a graph structure enhanced Transform encoder, inputting the sign language grammar structure into a Transform decoder integrating grammar consistency loss, and generating a target text conforming to target natural language grammar and semantic rules. According to the method, the problems of continuous sign language action adhesion and grammar structure difference are effectively solved, and the sign language translation accuracy and the natural language generation fluency are greatly improved.

The invention provides a method and device for generating an SQL instruction through a dialogue type natural language, and relates to the technical field of language processing.The method comprises the steps that a natural language query request input by a user is received; calling a neural symbol reasoning module to extract a query intention and a key entity list, generating a plurality of candidate SQL instructions through symbol reasoning, and evaluating confidence; if the highest confidence coefficient is lower than a set threshold value, multiple rounds of dialogue clarification are initiated and fed back to the neural symbol reasoning module; calling an adaptive SQL reconstruction engine to perform screening, reconstruction or optimization based on a dynamic data environment on the target candidate SQL instruction; and executing the final SQL instruction, obtaining a query result and returning the query result to the user. By means of the method and device, the technical problem that in the prior art, due to fuzziness of natural language query and dynamic complexity of a data environment, SQL instruction generation accuracy is low, and then query execution efficiency is affected is solved, and the query execution efficiency is improved while SQL semantic accuracy is guaranteed.

The invention discloses a marketing copywriting generation method and device and a storage medium, and relates to the technical field of natural language generation, and the method comprises the steps: obtaining a to-be-optimized original marketing copywriting, a structured tag and environment variable information; inputting the original marketing copywriting, the structured tag and the environment variable information into a causal reasoning generation model; in response to a virtual intervention instruction of a user on at least one target intervention label in the structured labels, executing anti-factual reasoning through a causal reasoning generation model, generating an optimized marketing copywriting corresponding to the virtual intervention instruction, and calculating a corresponding predicted interaction amount; and outputting the optimized marketing copywriting and the predicted interaction amount. According to the method, the anti-factual causal reasoning and the large language model generation capability are fused, the user is supported to preview and optimize the copywriting and the interaction effect in real time through the virtual intervention label, multi-strategy low-cost and efficient simulation evaluation is realized on the premise that real delivery is not needed, and the intelligent level, the decision-making efficiency and the result interpretability of copywriting optimization are remarkably improved.

Techniques are provided for producing an answer to a question regarding a domain. A natural-language textual sequence representing the question is received. From a knowledge graph associated with the domain, first and second textual passages are received using rankings corresponding to the natural-language textual sequence, a first textual summary is received summarizing textual information in a first vicinity of the first textual passage, and a second textual summary summarizing textual information in a vicinity of the second textual passage is received. An answer to the question is obtained using a language model by encoding a first intermediate output based on the natural-language textual sequence, the first textual passage, and the first textual summary, encoding a second intermediate output based on the natural language textual sequence, the second textual passage, and the second textual summary, and decoding a concatenation of the first and second intermediate outputs. An output is provided.

The invention relates to the technical field of fire coal purchase decision support, and particularly discloses a fire coal purchase decision method, system and equipment based on a big language model.The fire coal purchase decision method based on the big language model.The fire coal purchase decision method based on the big language model comprises the steps that target data are acquired; inputting the target data into the large language model, and outputting a fire coal purchase basic data analysis result; according to a fire coal purchasing basic data analysis result, further analysis is carried out through a large language model, and a fuel purchasing strategy direction suggestion is generated; according to a strategy direction selected by a user, constructing a multi-target optimization model, and solving the multi-target optimization model to obtain an optimal combination of purchased coal types, quantity and purchasing routes; according to the optimal combination of the types, the quantity and the routes of the purchased coal, a customized fire coal purchasing scheme is generated through the natural language generation capability of the large language model, the fire coal purchasing decision is intelligent, scientific and efficient, and powerful support is provided for reducing cost, improving efficiency and guaranteeing safety for thermal power enterprises.

The invention relates to the technical field of enterprise ESG digital management, in particular to a sustainable management ESG platform. The platform comprises a data access layer, a twin modeling layer, an intelligent analysis layer, a report generation layer and a collaborative interaction layer. The data access layer is used for standardized acquisition, cleaning and block chain evidence storage of multi-source ESG data; the twinborn modeling layer constructs enterprise ESG digital twinborn bodies through index mapping and scene simulation; the intelligent analysis layer realizes index dynamic monitoring, abnormity early warning and risk quantitative evaluation based on time sequence prediction and causal reasoning; the report generation layer outputs a compliance report through a multi-template automatic adaptation and natural language generation technology; and the collaborative interaction layer realizes multi-role data sharing, task tracking and strategy optimization. According to the method, multi-modal data fusion and an intelligent algorithm are fused, the efficiency of ESG data integration, risk response, report generation and collaborative management is comprehensively improved, and the problems of data isolation, response lag and manual dependence in a traditional mode are effectively solved.

The invention relates to a power grid multi-modal data compliance monitoring method based on a knowledge graph, and the method comprises the following steps: S1, obtaining multi-source heterogeneous data of a power grid, carrying out the preprocessing, and generating structured and semantic data representation; s2, constructing an initial knowledge graph according to the produced structured and semantic data representation; s3, according to the initial knowledge graph, combining a time perception mechanism TGN to identify compliance standards and laws of a power grid operation environment changing with time, combining rule-based logical reasoning with a deep neural model through neural symbol fusion, realizing compliance judgment under a fuzzy condition, and constructing a knowledge agent; s4, based on the knowledge agent, performing multi-modal compliance analysis in combination with deep learning and symbolic reasoning to obtain an analysis result; and S5, based on an analysis result, in combination with a natural language generation technology, automatically generating an interpretation report for the violation event. According to the invention, dynamic intelligent compliance analysis is realized, and the intelligent and automatic level of power grid management is significantly improved.

The invention discloses an automatic advertisement creativity generation method and system based on natural language generation. According to the invention, through establishing a keyword and context construction module, a brand semantic modeling and control module, an industry trend extraction module, an advertisement generation engine, a diversity control module and a putting adaptation and post-processing module, automatic generation of advertisement originality is realized. The system firstly receives a keyword and a mood instruction input by an advertiser, and constructs an initial semantic context; and then, coding a brand sample text by utilizing a pre-training model, and generating a brand semantic center vector as a prompt embedding so as to guide the generated content to accord with brand tonality. Meanwhile, through a keyword reservation supervision mechanism, it is ensured that core keywords appear naturally and at high frequency in the process of generating the copywriting. In the generation process, the system also uses a sampling strategy through a diversity control module and uses a semantic vector to remove duplication, and finally outputs diversified and non-redundant high-quality advertisement copywriting.

The invention belongs to the technical field of databases, and particularly relates to an LLM-Text2SQL-oriented database table relation exploration method. According to the method, a multi-stage cooperative processing strategy is adopted, database metadata and data content features are integrated, and semantic enhancement and structured completion are performed on original information through a large language model. And on the basis, a potential association candidate set is screened in combination with an algorithm based on feature similarity, multi-dimensional verification and judgment of an association relationship are performed by fusing a predefined rule and a large language model reasoning mechanism, and finally an entity relationship graph supporting interactive editing and iterative optimization is generated. According to the method, end-to-end automatic processing from a heterogeneous database with constraint missing and data integrity impaired to a standardized ER graph is realized, and a solid foundation is laid for remarkably improving the success rate of natural language SQL generation based on a large language model.

Some embodiments relate to receiving a natural-language textual sequence representing; retrieving, from a knowledge graph, a first textual passage and a second textual passage based on rankings with respect to the natural-language textual sequence, a first textual summary summarizing textual information in a first vicinity of the first textual passage, and a second textual summary summarizing textual information in a vicinity of the second textual passage; obtaining the textual output in response to the textual input using a language model by encoding a first intermediate output based on the natural-language textual sequence, the first textual passage, and the first textual summary, encoding a second intermediate output based on the natural language textual sequence, the second textual passage, and the second textual summary, and decoding a concatenation of the first intermediate output and the second intermediate output; and providing an output to a user based on the textual output.

The invention discloses an interactive personalized information pushing method, and the method comprises the steps: obtaining user data, building a dynamic user portrait for a user based on the user data, and enabling the dynamic user portrait to comprise a demand tag, a preference weight coefficient and an economic capability evaluation index; user requirements are analyzed in real time through natural language interaction, and dynamic features and intention keywords are extracted; in combination with multi-dimensional information of a commodity database, the system adopts a priority matching algorithm to generate a recommended commodity list, and generates personalized push content through a natural language generation engine, including commodity core selling points, matching degree analysis and interactive question and answer guidance; and finally, the system sends the content to a user terminal through multi-channel pushing, monitors user behaviors in real time to update a user portrait, and continuously optimizes a recommendation strategy.

The techniques described herein enhance the operations of natural language generation systems through training and / or augmentation by a large language model. In a first example, the large language model can execute training operations by processing a training dataset to produce a natural language output. The natural language generation system can analyze the training dataset and the natural language output to generate a natural language output mimicking the output of the large language model. The large language model can then evaluate the output of the natural language generation system to iteratively adjust and improve the quality of natural language outputs. In a second example, the large language can augment a small language model in executing natural language tasks. This is accomplished by retrieving external information using the large language model to generate an augmentation input to provide context and a language framework to the small language model to enhance overall outputs.

The invention relates to the technical field of power dispatching, and discloses an energy-carbon platform multi-source early warning information generation method, system and device based on a large language model, and a medium, and the method comprises the steps: collecting multi-source energy data of a monitoring region, and carrying out the data preprocessing; performing data fusion through a multi-modal data fusion method, and performing anomaly identification through an anomaly prediction model to obtain a predicted value and a composite anomaly score; and constructing a structured context based on the predicted value and abnormal associated data obtained by composite abnormal score analysis, and generating natural language early warning information by constructing a large language model. According to the method, by fusing multi-source data and combining the natural language generation capability and the numerical value verification mechanism of the large language model, intelligent identification and accurate description of abnormal conditions and automatic generation of executable suggestions are realized.

The invention provides a structure multi-type apparent damage identification and detection report generation method based on a multi-modal large model. According to the method, the image recognition result and a natural language generation technology are fused, so that full-process automation from damage detection to report generation is realized, and the intelligent level of building structure health monitoring is remarkably improved. According to the method, automatic identification and accurate positioning of multi-type damage of the appearance of civil engineering structures such as buildings and bridges can be realized, and objective and normalized maintenance suggestions and detection reports are generated in combination with a vision-language multi-modal modeling method, so that the intelligence and automation level of structural appearance damage detection and operation and maintenance management is improved.

The invention provides a fire monitoring method and system and computer equipment. The method comprises the following steps: acquiring a real-time image of a monitoring scene; calling a visual large model to analyze the real-time image, and generating fire description information of the real-time image through the visual large model under the condition that the visual large model detects that a smoke and fire target exists in the real-time image; and calling a large language model, and generating a rescue strategy according to the fire description information. According to the method, on one hand, the flame and smoke characteristics in the image can be identified by utilizing the capability of the visual large model, so that whether a fire occurs or not is accurately judged, and the false alarm rate can be effectively reduced; on the other hand, when it is judged that a fire occurs in the monitoring scene, fire description information is generated according to the real-time image. The fire description information serves as input of a large language model, and rescue suggestions are provided by means of natural language generation, inference analysis and knowledge association ability of the large language model. According to the method, the rescue action can be performed more quickly, and the rescue efficiency is improved.

The invention provides a geological disaster intelligent early warning and rapid report generation method based on a multi-modal large model, which is applied to the technical field of geological disaster prediction and comprises the following steps: acquiring geological monitoring data acquired by integrated geological disaster monitoring equipment; inputting the preprocessed image data into a deep convolutional neural network model to obtain image features; inputting the preprocessed time sequence monitoring data into a bidirectional long short-term memory network model to obtain a feature association relationship output by the bidirectional long short-term memory network model; performing multi-modal fusion based on the image features and the feature association relationship to obtain cross-modal joint features; inputting a preset influence factor and the cross-modal joint feature into the long short-term memory network model to obtain a disaster prediction result; and performing natural language generation on the disaster prediction result to obtain a structured report corresponding to the disaster prediction result. According to the invention, accurate prediction and rapid disaster report generation of geological disasters can be realized.

The invention discloses an AI policy consultation database and an AI consultation method, and belongs to the technical field of artificial intelligence and government affair information services. The invention aims to solve the problems of unreliable information source, large semantic understanding deviation and low policy matching accuracy of the existing general AI tool in the field of policy consultation. According to the technical scheme, the method comprises the steps that a multi-level, multi-dimension and versioning management policy database is constructed, and entity version snapshots and a multi-version concurrency control (MVCC) mechanism are adopted to process time sequence conflicts caused by policy revision; the retrieved trusted policy text and an original query request of a user are integrated into an enhanced Prompt (Augmented Prompt); and receiving and analyzing a preliminary conclusion returned by the LLM, and generating a standardized policy consultation report containing a structured table and attachments by using a natural language generation (NLG) and chapter structure analysis technology. And the accuracy, timeliness and reliability of policy consultation are remarkably improved.

The present invention relates to Smart visual sign language translator, interpreter, and generator. The invention is a process of two-way real time communication between person with hearing and / or speech disabilities (PwD) and normal person, wherein sign language of PwD (100) is captured as video, converted into image frames, interpreted using DCN trained model (104) and context based Natural language generation (105); and then converted into text (106) and speech (107) for a normal person to understand; similarly, communication of normal person is captured using mic (108), and is converted to text using speech to text converter (109) wherein speech recognition is done using Deep Stack Network. Module of Text and context analysis using Natural Language Understanding (110) analyses text received from (109). Interpreted text from (110) is used for predicting text to nearest sign image. These sign images are used for creating meaningful sequence of sign language gestures in (112).

The embodiment of the invention provides a form generation method and related equipment, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining form item configuration from a form item knowledge base; acquiring a natural language input by a user; generating a current question word based on form item configuration and a natural language; inputting the current question word into the large language model to obtain a form configuration JSON (JavaScript Object Notation) content; analyzing the form configuration JSON content through a JSON analyzer to obtain a current front-end page; in response to the first instruction, a current front-end page is displayed in the designer. According to the embodiment of the invention, the visual form configuration can be generated and previewed through the natural language, the user operation is reduced, the development threshold is lowered, and the development efficiency and the user experience are improved.

A multi-modal signal recognition and dialogue method based on a large language model comprises the steps that a signal coding module based on time sequence modeling is constructed, preprocessing and feature extraction are conducted on input I / Q signal data, and semantic alignment pre-training of signal features and modulation type text description is achieved through a contrast learning mechanism; designing a multi-modal fusion architecture, mapping signal features to a hidden space of a large language model by adopting a signal projector, and realizing deep fusion of the signal features and text features through special markers; constructing a dialogue generation module based on a pre-trained large language model, receiving the fused multi-modal input, and generating a natural language answer about signal analysis; performing feature alignment by training a double-layer MLP projector to realize end-to-end multi-modal signal understanding and dialogue ability; an intelligent question-answering system in a reasoning stage is constructed, natural language interaction between a user and the system is realized through a predefined professional prompt word template and a signal feature fusion mechanism, and multi-dimensional signal analysis query requirements are supported. According to the invention, the organic combination of signal understanding and natural language generation is realized, and the accuracy of signal identification and the user interaction experience are improved.

An embodiment may involve receiving a textual description of a workflow; determining, based on the textual description and operational characteristics associated with a computing system, a workflow outline that represents respective steps of the workflow; for each of the respective steps of the workflow, identifying a respective portion of the operational characteristics based on determining that the respective portion satisfies a relevance criterion; obtaining, from a natural language model and based on the respective portions of the operational characteristics, a respective programmatic representation for each of the respective steps of the workflow; and updating the workflow outline based on the respective programmatic representations.