Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

35 results about "Automatic summarization" patented technology

Automatic summarization is the process of shortening a text document with software, in order to create a summary with the major points of the original document. Technologies that can make a coherent summary take into account variables such as length, writing style and syntax.

Substation detection report automatic generation method based on artificial intelligence

The invention discloses a transformer substation detection report automatic generation method based on artificial intelligence, and the method comprises the following steps: obtaining primary detection data through multi-modal data collection and preprocessing; performing standard verification on the primary detection data by using a rule engine, completing trend analysis and anomaly detection, and generating structured detection data; inputting the structured detection data into an automatic summarization and suggestion generation module to obtain an analysis abstract and a professional suggestion; filling the structured detection data, the analysis abstract and the suggestion into a standardized detection report template for automatic matching; a standard detection report is generated by adopting automatic typesetting and is pushed to a multi-role cooperation module, and the standard detection report is audited by related auditing personnel; and after checking and confirming, filing the detection report, and distributing the detection report to a related management system or a specified terminal according to a set rule. According to the invention, intelligent acquisition, anomaly detection, automatic report generation and whole-process closed-loop management of multi-modal data are realized, and the detection efficiency, accuracy and report standardization level are greatly improved.
Owner:GANSU TRANSMISSION & DISTRIBUTION ENG CO

System and method for automatic summary generation

In accordance with the present disclosure, one or more computing devices and / or methods are provided. In an example, a query may be received from a client device. In response to the query, a plurality of search results corresponding to a plurality of internet resources associated with the query may be generated. A plurality of content items may be generated based upon the plurality of search results. The plurality of content items may be generated using a plurality of content extraction tools. A language model may be used to generate a plurality of summaries of the plurality of content items. Summary scores of the plurality of summaries may be determined. A first summary may be selected from the plurality of summaries based upon the summary scores. In response to selecting the first summary, the first summary may be provided for display on the client device.
Owner:YAHOO ASSETS LLC

Chinese long text-oriented hybrid automatic abstract generation method

The invention discloses a Chinese long text-oriented hybrid automatic abstract generation method, and relates to the technical field of natural language processing. According to the method, an extraction type abstract method in an unsupervised mode is taken as a first stage, the complexity of a model is reduced, meanwhile, information loss is avoided, and long text content is compressed. Then, a generative abstract module is constructed based on comparative learning, document-level coding representation containing context semantic information is fused into the training process of the model, so that the model effectively guides the generated abstract to be compared with a standard abstract text, irrelevant information is prevented from being generated, and the problem of exposure deviation generally existing in the generative text abstract is solved; and the abstract generation quality is improved while the model complexity is reduced.
Owner:SHENYANG AEROSPACE UNIVERSITY

Cross-file information summarization and new knowledge automatic summarization method and system

The invention discloses a cross-file information summarization and new knowledge automatic summarization method and system, and belongs to the technical field of file data processing, and the method specifically comprises the steps: collecting document file data, vectorizing the document file data based on a semantic embedding model, carrying out the reconstruction and semantic clustering of the semantic vector of the document file data, and carrying out the automatic summarization of the new knowledge. The method comprises the following steps: constructing a semantic structure chart which represents a logical relationship between document file data, performing content completion on nodes which are not completely associated in the semantic structure chart, extracting structured knowledge units from the completed semantic structure chart, and generating a new knowledge text based on the structured knowledge units; according to the method, the limitation of low efficiency of knowledge splitting and manual summarization between traditional document files is solved, the automation degree of knowledge discovery is remarkably improved, and the method is suitable for scenes of knowledge base construction, domain rule extraction, domain knowledge discovery and the like.
Owner:GUIZHOU BLUE DREAM FACTORY TECH CO LTD

Tibetan long text automatic abstract generation method based on two-stage structure

The invention provides a Tibetan long text automatic abstract generation method and system based on a two-stage structure, a storage medium and electronic equipment, and relates to the field of natural language processing. According to the method, firstly, a heterogeneous graph structure is constructed based on a Tibetan long text to be processed; secondly, taking the heterogeneous graph structure as input of a pre-trained graph neural network, executing multiple rounds of information interaction among nodes, obtaining enhanced sentence node representation in combination with an attention mechanism, and screening out a key sentence set by utilizing a binary classifier; and finally, based on the key sentence set, generating a text abstract of the Tibetan long text. According to the scheme, a long text can be better processed, the global semantic modeling capability is enhanced, and the text abstract generation method with low-resource language characteristics is considered, so that the accuracy, the continuity and the generalization capability of the abstract are improved.
Owner:MINZU UNIVERSITY OF CHINA

Task allocation and opinion filling method based on inspection outline

The invention belongs to the technical field of power transmission and transformation project cost effect supervision and inspection, and particularly relates to a task allocation and opinion filling method based on an inspection outline, which comprises a task allocation and opinion filling system. The task distribution and opinion filling system comprises a check outline dynamic analysis module, a check key point one-by-one filling module and a check key point automatic summarization module. The inspection outline dynamic analysis module is electrically connected with the inspection key point one-by-one filling module, the inspection key point one-by-one filling module is electrically connected with the inspection key point automatic summarizing module, and the inspection outline dynamic analysis module is used for forward construction of an inspection outline and reverse analysis of the inspection outline. The device solves the problems of file complexity, unfixed auditing experts, crossed auditing content, need of a large amount of manpower in later summarization, prolongation of the project period and increase of the working difficulty to a certain extent in the current settlement inspection work of the power transmission and transformation project.
Owner:STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO

Proactive alert aggregation and correlation management with automated summarization

An information handling system includes a data store configured to store an account associated with a first alert and a second alert. A processor may receive the first alert and the second alert, map the first alert and the second alert to a first node and a second node of a causality graph, traverse the causality graph starting from the sink node to the source node to determine an association between the first alert and the second alert, and generate the account based on the association between the first alert and the second alert.
Owner:DELL PROD LP

Automatic summarization method and system for legal documents based on semantic segmentation

The application discloses a kind of legal document automatic abstract method and system based on semantic segmentation, belong to natural language processing field.The application obtains civil first instance adjudication document as input, using the method of continuous sentence classification, the semantic segmentation is carried out to adjudication document, and the text paragraph of five parts, such as dispute category, plaintiff claim, defendant statement, fact and reason, judgment basis, judgment main text and tail portion is divided into;Using the method of generative text abstract is obtained abstract for each text paragraph after segmentation;For the abstract generated by each segmented paragraph of the same adjudication document, the final result is obtained by splicing in order.The application carries out automatic abstract to legal document, using the method of semantic segmentation, shortens the text length of single input generation abstract model, and can retain complete original text semantic structure features.
Owner:ZHEJIANG UNIV

Method for automatically summarizing material disposal and checking asset valuation

ActiveCN122022698BFeature vectorData set
The application relates to the technical field of asset valuation, and discloses a material disposal condition automatic summarization and asset valuation verification method, which comprises the following steps: acquiring attribute feature vectors of newly-warehoused materials, screening a reference category set from a mature material category library, fusing a multivariate regression model parameter set of the reference category to generate a prior parameter distribution, extracting limited historical records of the newly-warehoused materials to generate a multivariate sample data set, generating a posterior regression parameter distribution based on Bayesian fusion, calculating a predicted residual value rate and a predicted uncertainty, generating a predicted valuation, comparing the predicted valuation with a declared valuation, and outputting a verification result, so that the problem of insufficient accuracy of multivariate residual value prediction under a small sample condition is solved.
Owner:CHINA TOWER CO LTD

Material disposal condition automatic summarization and asset estimation verification method

The invention relates to the technical field of asset valuation, and discloses a material disposal condition automatic summarization and asset valuation verification method, which comprises the following steps of: obtaining attribute feature vectors of new warehousing materials, screening a reference category set from a mature material category library, and fusing a multiple regression model parameter set of a reference category to generate prior parameter distribution; limited historical records of new warehousing materials are extracted to generate a multivariate sample data set, posterior regression parameter distribution is generated based on Bayesian fusion, a prediction residual value rate and prediction uncertainty are calculated, a prediction estimated value is generated and compared with a declaration estimated value, a verification result is output, and the problem that multi-factor residual value prediction is insufficient in accuracy under the small sample condition is solved.
Owner:CHINA TOWER CO LTD

Intelligent medical literature searching and question-answering system

The invention discloses an intelligent medical literature search question-answering system, which belongs to the technical field of artificial intelligence, and realizes accurate understanding of a user query intention, efficient retrieval of medical literatures, automatic summarization of literature contents and real-time generation of a dialogue outline through natural language processing and deep learning technologies. The system interacts with the user through a multi-round dialogue mechanism to gradually refine and confirm the requirements of the user and provide more accurate literature search results and professional suggestions.
Owner:李若泰

Business Text Summary Generation Method and System Based on an Intelligent Platform

Embodiments of the present application relate to the field of artificial intelligence technology, and specifically to a method and system for generating business text summaries based on a smart platform. Obtain the smart government business conversation text stream; mine the business interaction state knowledge vector chain from the smart government business conversation text stream according to the node set mined from the business interaction state; based on the business element heat evaluation data of each business interaction state knowledge vector in the business interaction state knowledge vector chain, determine the state knowledge confidence of each business interaction state knowledge vector; based on the state knowledge confidence, determine the target key conversation text semantics from the business interaction state knowledge vector chain. The smart government platform system can realize the automatic summary generation function of the online business conversation to be processed, thereby improving the efficiency and quality of government affairs processing.
Owner:GUANGZHOU PINGYUN INFORMATION TECH CO LTD

Automated summary generation of content

Communications may occur within multiple different communication channels (e.g., email messages, chat messages, text messages, audio calls, messages exchanged via a website) and be stored within sources associated with the communication channels. A request to summarize the communications may be received via a user interface. One or more models may extract such interaction data, aggregate the interaction data, and process the interaction data to generate a summarization of the interaction data. The summarization may be presented via the user interface for subsequent review and approval. The summarization may be stored in association with a case associated with the communications.
Owner:SALESFORCE INC

An event element automatic abstract generation method and system based on a large language model

The application provides an event element automatic abstract generation method and system based on a large language model, and belongs to the technical field of environmental monitoring. The application obtains abnormal discharge, water quality and operation state records of a water supply and drainage pipe network in a target period; determines an effective time period, a connection relationship, an isolation relationship and a medium conveying direction according to the operation state records, and arranges event records into a water supply event record sequence and a drainage event record sequence; generates an event element generation instruction including a pollution source, a pollution invasion site, an affected water supply object, a disposal matter and a corresponding event record identifier for the water supply and drainage connection relationship; inputs the event record sequence and the event element generation instruction into a large language model to generate a cross-contamination event element, and inputs the cross-contamination event elements of the same water supply and drainage connection relationship into the large language model to generate a cross-contamination event abstract. The application can improve the association integrity of monitoring information and operation and maintenance events and the abstract accuracy.
Owner:北京清研灵智科技有限公司

System

To provide diversity of information and easiness of access by daily generating a newspaper released from an era and geographical restrictions by a system using generative AI.SOLUTION: A system for daily generating a newspaper released from an era and geographical restrictions by using generative AI includes information collecting means, automatic summarizing means, and article generating means.SELECTED DRAWING: Figure 1
Owner:SOFTBANK GROUP CORP

System and method for automatic summarization in interlocutor turn-based electronic conversational flow

A computer-implemented method and computer system improvements for automatically generating a summary note related to a digitally-recorded interlocutor conversation, such as a chat transcript, including accessing a data corpus having at least one digitally-recorded conversation of text-based interlocutory conversation(s) including at least one labeled value, extracting at least one Conversation Feature from the labeled value(s), creating at least one Summary Feature from the extracted Conversation Feature(s), generating at least one Summary Note by combining one or more Narrative Structures with the Summary Feature(s), and digitally outputting the at least one Summary Note.
Owner:DISCOURSE AI INC

Semantic understanding and automatic abstract generation system for literature works

The invention relates to the field of artificial intelligence, in particular to a semantic understanding and automatic abstract generation system for literature works, which comprises a text collection module, a semantic understanding module, a theme extraction module, an abstract generation module and an optimization module. Performing semantic analysis on the literature data set in combination with hierarchical aggregation, multi-level information fusion and a knowledge graph, performing deep pre-training on text data through a RoBERTa model, and capturing a complex semantic relationship in the literature; the method comprises the following steps: pre-training a semantic understanding model by adopting a shielded language model, optimizing the shielded language model by adopting a DTO-PSO hybrid optimization algorithm, and introducing an information entropy feedback mechanism and a constraint processing mechanism to improve the depth and accuracy of semantic understanding.
Owner:SICHUAN MEDIA COLLEGE

Archive text automatic abstracting and knowledge question-answering system combined with large model

The invention discloses an archive text automatic abstracting and knowledge question-answering system combined with a large model, and relates to the technical field of text automatic abstracting and question-answering. Comprising a text scanning module for performing text scanning on a target archive text, and extracting text structured data through an entity recognition model to obtain text information items; and the information screening module is used for carrying out information screening on the basis of the text information items, screening the information into judgment data and non-judgment data, taking the judgment information as input data, and generating matching information corresponding to the input data through a large language model. According to the method, the file noise interference is resisted through a triple mechanism, the cross-modal alignment technology performs joint analysis on the OCR text, the seal image and the table data, so that the fuzzy character recognition accuracy is improved, the mapping relation between the key information item and the alignment information item is established, and the spelling dislocation problem of the digital file is solved.
Owner:SICHUAN VOCATIONAL & TECHN COLLEGE

Simple daily filling and data analysis method and system

The invention provides a simple daily report filling and data analysis method and system, and relates to the technical field of data management.The method comprises the steps that daily report data submitted by a user through a front-end customized filling interface is received, and the daily report data is structured daily report data formed based on a preset department business field template; associating the structured daily report data with the user and department information through an object relation mapping framework, and storing the structured daily report data and the user and department information in a preset database to form an associatively stored daily report data set; based on the associatively stored daily report data set, performing automatic summarization analysis through a preset multi-dimensional statistical rule, and generating a quantitative statistical result; and the quantitative statistical result is converted into a visual chart to be displayed so as to support management decision making. According to the method, the problems of difficulty in daily report data aggregation calculation, visualization deficiency and insufficient decision support in the prior art can be solved.
Owner:SHANGHAI TONGLIAN FINANCIAL SERVICES CO LTD

Intelligent surveillance and automatic summarization method

The present invention relates to an intelligent surveillance and automatic summarization method for generating a summary video by automatically determining a summarization segment of an original image. The intelligent surveillance and automatic summarization method comprises the steps of: (a) receiving an original image to detect objects and storing information about the detected objects; (b) calculating a congestion level of the objects in the original image to determine a segment for summary video generation in the original image; and (c) generating and storing a summary video according to the determined segment.
Owner:ZIOVISION CO LTD

An automatic summarization method and system based on the fusion of TextRank and multi-dimensional semantic features

The invention discloses an automatic summarization method and system based on the fusion of TextRank and multi-dimensional semantic features. The method extracts news keywords, obtains news topics, and extracts sentence feature information and vector representation of sentences from pre-processed news. Four factors, namely, sentence-topic similarity, sentence-title similarity, keyword coverage, and whether a feature word is contained, are used as factors affecting sentence weights. These factors and the original TextRank score are applied to the automatic summarization in an optimized combination. The TextRank score and the sentence feature score are added and averaged to obtain a sentence comprehensive score. The MMR algorithm is used to calculate the similarity between sentences to remove sentences with high redundancy. Finally, the first n sentences are taken as summaries according to a set number according to the similarity sorting. The efficiency of removing redundant sentences by the original formula is improved, the summary content of the news can be accurately obtained, and the accuracy of the news content overview is greatly improved, thereby helping people reduce reading time.
Owner:XIAN TECH UNIV +1

Text summarization generation method based on keyword information and multi-head attention mechanism

The application relates to the technical field of natural language processing, and discloses a text abstract generation method based on keyword information and a multi-head attention mechanism, which takes a classical seq2seq network model with an attention mechanism as a main body, and adds a keyword acquisition branch, the keyword acquisition branch is used for keyword extraction and filtering on input text to acquire refined keywords, and the refined keywords are input into an attention mechanism module of the classical seq2seq network model for subsequent abstract generation; Chinese texts are collected, a data set is constructed, an abstract generation model is trained, and then the trained abstract generation model is used for automatic abstract generation on to-be-inspected texts. The method has a remarkable effect on the improvement of the text abstract effect, and the abstract effect example shows that the model has good readability and pays attention to keyword information of the original text.
Owner:SHANGHAI NORMAL UNIVERSITY

An event extraction method based on a machine reading comprehension model

The present invention discloses an event extraction method based on a machine reading comprehension model. The method models the two subtasks of event extraction: event detection and event element extraction, into machine reading comprehension tasks, and uses a multi-round question-answering approach to extract complete information about the event. The present invention constructs a machine reading comprehension framework by using a BERT pre-training model. In order for the model to learn the entity information in the sentence, the entities in the sentence are explicitly marked in the input sentence. In addition, a historical conversation information encoding module is constructed, and an attention mechanism is used to capture important information from historical conversations to assist in the current round of question-answering. Compared with existing methods, the method of the present invention can better utilize prior knowledge to assist in event extraction, and has broad application prospects in automatic summarization, automatic question-answering, and the construction of causal graphs.
Owner:SICHUAN UNIV

System

To provide a system.SOLUTION: A system is configured to efficiently improve business operations in a remote-work-centered environment, the system being configured by integrating online conference means, chat communication means, automatic summarization means, task management means, and document sharing means.SELECTED DRAWING: Figure 1
Owner:SOFTBANK GROUP CORP

Business text abstract generation method and system based on intelligent platform

The embodiment of the invention relates to the technical field of artificial intelligence, in particular to a business text summary generation method and system based on an intelligent platform, and the method comprises the steps: obtaining an intelligent government affair business session text flow; mining a service interaction state knowledge vector chain from the intelligent government affair service session text flow according to a service interaction state mining node set; based on the business element thermal evaluation data of each business interaction state knowledge vector in the business interaction state knowledge vector chain, determining the state knowledge confidence of each business interaction state knowledge vector; based on the state knowledge confidence, the intelligent government affair platform system determines the target key session text semantics from the business interaction state knowledge vector chain, and can realize an automatic abstract generation function for the online business session to be processed, thereby improving the efficiency and quality of government affair processing.
Owner:GUANGZHOU PINGYUN INFORMATION TECH CO LTD

Automatic Abstract Generation Method Based on Pointer Key Information

The present invention discloses an automatic abstract generation method based on pointer key information. First, sentence vectors of multi-dimensional semantic features of an article are obtained through a pre-trained model. Secondly, pointer is used to select keyword semantic feature word vectors. Finally, an encoder is used to generate an abstract. In this method, a model framework is built using a neural network, and automated learning and training are used, eliminating excessive manual intervention. The training process adopts two methods of prediction and verification, and then the model calculates the loss and automatically improves the accuracy of the model. The method of the present invention has strong applicability and will have great applications in aspects such as news title generation and copywriting generation.
Owner:GUILIN UNIV OF ELECTRONIC TECH

An Event Extraction Method Based on Thematic Features and Implicit Sentence Structure

ActiveCN113901813BSemantic analysisEnergy efficient computingSingle sentenceSyntax
The present invention discloses an event extraction method based on topic features and implicit sentence structures, which is mainly used to present unstructured texts containing event information in a structured form and has wide applications in fields such as automatic summarization, automatic question answering, and information retrieval. First, the present invention combines BERT and LDA to obtain the topic features of the document and introduce document-level topic information into the sentence-level event extraction model; secondly, the syntactic information implicit in the BERT word embedding representation is extracted, and the extraction process is jointly modeled with event extraction, introducing important syntactic information for event extraction while avoiding the problem of error accumulation; finally, the model uses a sequence labeling method based on Bi-LSTM and cascaded CRF to extract multiple trigger words in a single sentence and extract the element roles of entities in multiple events.
Owner:SOUTHEAST UNIV

Scientific literature automatic summarization agent, system, and method based on large language model

PendingCN123031750ALinguistic modelEngineering
The application discloses a scientific research literature automatic summarization agent, system and method based on a large language model, comprising a literature access and preprocessing module, which pre-processes input pdf format literature and outputs a standardized text string; a skill-based workflow arrangement module, which synthesizes a calling request object from the standardized text string; an Agent runtime kernel module, which executes a two-stage reasoning strategy on the large language model calling request object, realizes skill-based knowledge injection, and drives flow event distribution; a multi-layer progressive JSON analysis and automatic repair module, which generates a structured summary object and three output formats of Markdown, HTML and PDF; and an SSE flow transmission and front-end rendering module, which outputs a seven-block visual summary view rendered by a web front end and export files in three formats of Markdown / HTML / PDF. When reading scientific research literature, knowledge extraction is more accurate and standardized, summary is hierarchical and adapted to multiple scenes, and the process is automatically and efficiently handled.
Owner:SHENZHEN UNIVERSITY OF ADVANCED TECHNOLOGY +1

Automatic checking method for budget demand data, computer equipment and storage medium

The invention relates to the technical field of data verification, and discloses a budget demand data automatic verification method, computer equipment and a storage medium, and the method comprises the steps: creating an Excel form comprising a data verification module, an articulation verification module and an automatic summarization calculation module; a data verification module is used for verifying budget demand data in the Excel form; an articulation checking module is used for checking the articulation between a general table and a detailed table in the Excel form; and automatically summarizing and calculating the budget demand data and the basic data of each unit in the Excel form by using an automatic summarizing and calculating module, and storing the data in a database. By introducing an advanced automatic verification tool, data can be more effectively integrated, processed and analyzed, so that the workload of manual operation is greatly reduced, the burden is reduced, the data processing and analysis time is shortened, the working efficiency is improved, and the resource allocation process is more efficient, accurate and reliable.
Owner:GUANGDONG POWER GRID CO LTD DONGGUAN POWER SUPPLY BUREAU

An automatic summarization factuality enhancement method based on syntax structure pruning

The application is an automatic summarization fact enhancement method based on syntax structure reduction, and relates to big data resource service and data mining technology.The method comprises the following steps: analyzing the syntax structure of a to-be-corrected abstract text to obtain a corresponding syntax dependency tree; performing postorder traversal on the syntax dependency tree, attempting to prune a sub-tree according to a syntax dependency relationship, obtaining a candidate abstract, and restoring the pruned sub-tree before sequentially accessing a next sub-tree; pre-deploying multiple fact evaluation index models on a server; calling the multiple fact evaluation index models on the server in real time to evaluate the fact of each candidate abstract; when a next abstract is better than a previous abstract on all fact evaluation indexes, the next abstract is considered to be better, and the best abstract is output.The application first applies syntax structure reduction to a post-correction scheme, greatly expands the range of errors that can be coped with, improves the fact accuracy of an automatic summarization model in generating an abstract, and can be widely used in existing automatic summarization models.
Owner:BEIJING UNIV OF POSTS & TELECOMM