20 results about "Citation" patented technology

A user provides a question to be answered from detailed, dense or otherwise complex documents to a processing system that converts the question to a structured query language query and generates an embedding from the question, augmented by temporal data, synopses, themes, or other relevant information or data. The embedding is compared to embeddings generated from documents of a knowledge base to identify documents that are relevant to the question, and to rank such documents for their relevance. Highly ranking documents are combined with the query and provided to a language model that returns an answer to the question. A source for the answer is identified in at least one of the documents. The answer and the identified documents are presented to the user.

The invention discloses a standard reference proofreading method and device based on a large language model and related equipment, and the method comprises the steps: obtaining a to-be-proofreading document, inputting the large language model, determining standard reference data in the to-be-proofreading document, and for each standard reference data, judging whether a consistent reference standard exists in a standard database or not; if yes, the reference standard consistent with the standard number of the standard reference data is called, if not, the reference standard is called from the standard database by means of large language model semantic comparison, each standard reference data is compared with the called reference standard to obtain a comparison result, and correction suggestions and explanations are generated based on the comparison result. According to the method and the device, the standard reference data in the to-be-proofreading document is determined by utilizing the large language model, the reference standard corresponding to the standard reference data is determined by adopting multi-layer verification logic and utilizing the standard number and the standard name, the comparison result is obtained through comparison, and the suggestion and the description are generated, so that standard reference proofreading is realized, and the accuracy of standard reference proofreading is improved.

The application relates to the technical field of deep learning and information processing, and proposes a citation information entity extraction method, device, equipment and medium, wherein the method comprises the following steps: acquiring sample data set of citation data; training a preset deep learning citation block model by using the sample data set to obtain a trained target citation block model; training a preset deep learning citation classification model by using citation block data output by the target citation block model to obtain a trained target citation classification model; acquiring original citation data to be processed, and processing the original citation data to be processed by using the target citation block model and the target citation classification model to extract citation information entities from the original citation data to be processed and obtain target citation entity item results; and outputting the target citation entity item results. According to the technical scheme, the accuracy and efficiency of citation information entity extraction are improved.

The application belongs to the field of text processing, and particularly relates to a citation annotation method, aiming to solve the problem of poor recognition effect of citation missing phenomenon. The method comprises the following steps: segmenting a to-be-annotated text into multiple subtexts and determining the position identifier of any subtext; in the case that a citation annotation symbol is recognized in any subtext, determining that the citation type identifier thereof is a first identifier; in the case that any subtext matches a citation description set comprising multiple citation description structures successfully, determining that the citation type identifier thereof is a second identifier, the citation description structure being obtained by screening from alternative citation description structures, the alternative citation description structure being obtained by performing citation description structure recognition on the subtexts of multiple to-be-annotated texts; and performing citation annotation in the to-be-annotated text based on the citation type identifier and the position identifier of each subtext. The application can annotate all citation information of the text, and improve the efficiency and intelligent level of citation annotation in various application scenarios.

This invention discloses a citation verification method, apparatus, device, medium, and computer program product, comprising: acquiring original input; extracting the claim text to be verified and a candidate citation list associated with the claim text from the original input to form structured citation information and opinion citation pairs; performing existence verification on each candidate citation in the candidate citation list based on the structured citation information to determine whether the target document corresponding to the candidate citation actually exists and is accessible, generating an existence verification result, which at least includes a reason code and / or a hit method; and performing semantic alignment and support determination on the claim text and evidence fragments in the target document based on the opinion citation pairs to generate an opinion verification conclusion, which at least includes suggested actions and / or evidence location information. These technical features automate the entire citation verification process, improving the accuracy, traceability, and engineering reusability of citation verification.

The invention belongs to the field of text processing, particularly relates to a citation labeling method, and aims to solve the problem of poor citation missing phenomenon recognition effect. The method comprises the following steps: segmenting a to-be-labeled text into a plurality of sub-texts and determining a position identifier of any sub-text; determining a reference type identifier as a first identifier under the condition that a reference tag is identified in any sub-text; under the condition that any sub-text is successfully matched with a reference description set comprising a plurality of reference description structures, the reference type identifier of the sub-text is determined as a second identifier, and the reference description structures are obtained by screening alternative reference description structures; the alternative reference description structure is obtained by performing reference description structure recognition on sub-texts of the to-be-labeled text; and based on the reference type identifier and the position identifier of each sub-text, performing reference labeling in the to-be-labeled text. All quotation information of the text can be labeled, and the quotation labeling efficiency and the intelligent level under various application scenes are improved.

The application discloses a text evaluation method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining a text to be evaluated; determining all citation fields in the text to be evaluated and the literature identification numbers corresponding to the citation fields according to the citation marks in the text to be evaluated; determining at least one text attribute and at least one citation attribute of the text to be evaluated according to the citation fields and the corresponding literature identification numbers; and evaluating the text to be evaluated according to the text attributes and the citation attributes. In the technical scheme of the embodiment of the application, different text attributes and different citation attributes can represent the quality of the text to be evaluated from different dimensions, and the text to be evaluated is comprehensively evaluated, so that the accuracy of the text evaluation can be improved.

A system, method and non-transitory computer readable medium for organizing scientific and technological publication data for performing predictive analysis is provided. According to embodiments of this disclosure, organizing scientific and technological publication data for performing predictive analysis includes: receiving the publication data from a publication database; parsing the publication data to obtain citations and associated citation data for each publication; storing the citation data in a first database; identifying one or more co-citations between publications in the citation data to obtain co-citation information between publications; and storing the co-citation information between publications in a second database.

The present invention discloses a method for determining the influence of a paper, comprising: obtaining the full text of the citing documents of the paper to be analyzed, performing data preprocessing on the citing documents to obtain the citation texts in the citing documents, and annotating some of the citation texts to obtain a label category set for the citation texts; using the label category set of the citation texts obtained in step 1 as a training set to train a first DeepMoji model, and using the trained first DeepMoji model to perform sentiment classification on unannotated citation texts; obtaining social media mentions of the paper to be analyzed, repeating the above steps to obtain a trained second DeepMoji model, and using the second DeepMoji model to perform sentiment classification on unannotated social media mentions; and calculating the paper's influence index by assigning different weights to different emotions in the citation texts and social media mentions. The present invention achieves a more fine-grained influence calculation method that can distinguish different citation content and social media mention content.

According to aspects of the disclosed subject matter, systems and methods for providing a generated response to an input prompt are presented, where the generated response includes auditable, specific citations to one or more content sources. Moreover, and in various embodiments, the generated responses may utilize, in whole or in part, user-supplied content and / or user-identified content as content sources for responding to an input prompt.

The invention belongs to the field of citation processing, particularly relates to a Chinese citation type indexing method based on feature words and structures, and aims at solving the problems that existing citation classification depends on citation literature data, does not conform to the national standard, is poor in classification effect and depends on manual judgment. The method comprises the following steps: judging whether a literature carrier code exists in a Chinese quotation of a to-be-indexed type or not; if the quotation exists, segmenting the quotation into source information, text information and self-defined types by taking the literature carrier code as a demarcation point; if not, segmenting the Chinese quotation of the type to be indexed based on preset punctuation marks, and segmenting the quotation into source information and chapter information in combination with own features and feature words of the quotation; extracting keyword features in the chapter information and dictionary features related to literature types; and obtaining the citation structure characteristics in the Chinese citation of the type to be indexed, and judging the citation type through a predefined classification rule. Through feature extraction, the method does not need to depend on manual classification, and improves the classification precision.

A method, system, and computer program product is provided for tokenizing document citations. The method may include tokenizing each string of a plurality of strings into at least one citation token representing at least one citation parameter, resulting in a plurality of citation tokens, grouping the plurality of citation tokens into a plurality of token groups, each token group of the plurality of token groups including at least one citation token representing a separate citation in the at least one textual document, assigning metadata to each token group of the plurality of token groups, and generating a normalized citation for each token group of the plurality of token groups based on the metadata.

The invention discloses a text evaluation method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining a to-be-evaluated text; determining all reference fields in the to-be-evaluated text and literature identification numbers corresponding to the reference fields according to the reference marks in the to-be-evaluated text; according to each reference field and the corresponding literature identification number, determining at least one text attribute and at least one reference attribute of the to-be-evaluated text; and evaluating the to-be-evaluated text according to each text attribute and each reference attribute. According to the technical scheme, different text attributes and different reference attributes can represent the quality of the to-be-evaluated text from different dimensions, the to-be-evaluated text is comprehensively evaluated, and the accuracy of text evaluation can be improved.

The invention discloses an interpretable quotation recommendation method and system based on heterogeneous graph comparative learning. The method comprises the following steps: constructing an academic heterogeneous graph; extracting a reference relation view and an author cooperation relation view from the academic heterogeneous graph, and performing node embedding learning on the reference relation view and the author cooperation relation view by adopting a multi-view graph attention network to obtain embedded representation of the paper under the reference relation view and the author cooperation relation view; aligning the embedded representation of the same paper under the reference relationship view and the author cooperation relationship view through cross-view comparison learning; a multi-channel consistent weighted network is adopted to fuse embedded representation under the reference relation view and the author cooperation relation view, and popularity weight is introduced to reduce popularity deviation; and performing quotation recommendation prediction for the target paper based on the fused embedded representation, and generating an interpretable text for a quotation recommendation result. According to the method, the recommendation diversity and fairness are improved, and the interpretability of the recommendation system is remarkably improved.

The present invention discloses a method for automatic classification of citation texts based on a large language model, comprising: obtaining the full text of the cited document to be analyzed and the full text of its citing document, and performing data preprocessing on them; using the large language model to extract key information from the cited document; inputting the extracted key information and the obtained citation text information into a BERT model to extract corresponding features, aggregating the extracted features and the citation position features obtained in the data preprocessing, and using the aggregated features to train a multi-layer perceptron classification model for the classification task; after extracting features from the citation text to be classified through the above steps, aggregating the features and inputting them into the trained multi-layer perceptron classification model to predict the citation object label and citation function label of the citation text, and classifying the citation text according to the citation object label and citation function label. The present invention greatly improves the accuracy of citation text classification.

According to aspects of the disclosed subject matter, systems and methods for providing a generated response to an input prompt are presented, where the generated response includes auditable, specific citations to one or more content sources. Moreover, and in various embodiments, the generated responses may utilize, in whole or in part, user-supplied content and / or user-identified content as content sources for responding to an input prompt.

The application provides a scientific and technological literature innovation evaluation method based on a large model Agent technology, comprising: hierarchical classification of each citation source in a citation dataset in disciplines and research branches, determination of the discipline field code and specific research branch code thereof according to a discipline classification system, and obtaining citation distribution data containing complete classification labels; statistics of the number and proportion of citations of each research branch in the citation distribution data, and calculation of the crossing degree of the literature among different research branches; the large model Agent comprehensively evaluates the innovation value of the literature based on the crossing degree among the branches and the total number of citations, adjusts the weight of different literature according to the crossing degree among the branches, and generates an innovation value score dataset; the large model Agent cross- validates the candidate literature list, confirms the effectiveness of the cross-disciplinary breakthrough characteristics in combination with the information of the proportion of the number of citations of each branch, completes the final labeling, and identifies the breakthrough innovative literature.

This invention discloses a method and system for quantitatively assessing citation impact through sentiment semantic deconstruction, relating to the field of text analysis technology. The method includes: constructing a feature word score table; performing dependency parsing on the cited sentence to obtain a part-of-speech position feature set, segmenting the cited sentence, obtaining multiple cited clauses, removing non-evaluative feature words, and obtaining multiple optimized clauses; identifying the subject scores of the multiple optimized clauses and obtaining multiple subject score identification results; identifying the auxiliary word scores of the multiple optimized clauses and obtaining multiple auxiliary word score identification results; obtaining the score of each optimized clause and calculating the quantitative score of citation impact. This invention solves the technical problem in existing technologies where citation sentiment analysis only reaches a coarse-grained classification of positive, negative, and neutral, making it difficult to quantitatively assess the degree of citation impact. It achieves the technical effect of transforming qualitative citation impact evaluation into standardized quantitative scoring, improving the objectivity and comparability of the assessment.

The invention provides a reference type analysis method and system based on joint multi-task learning, and relates to the technical field of semantic classification, and the method comprises the steps: configuring input data and label data of a reference intention classification task; defining a reference evaluation classification task, and configuring input data and label data of the reference evaluation classification task; constructing a combined multi-task learning framework, and embedding the quotation intention classification task and the quotation evaluation classification task as parallel classification tasks into the combined multi-task learning framework; and adopting an alternate iteration strategy of parallel classification tasks, training a joint multi-task learning framework, balancing gradient propagation between parallel tasks, obtaining a reference type joint analysis model, and carrying out reference type analysis processing. By means of the method and device, the technical problem that in the prior art, deep semantic association between the citation intention and citation evaluation cannot be effectively mined, and the intelligent analysis ability of scientific research content is affected can be solved, and the technical effect of improving the interpretation ability of citation semantics is achieved.

A system, method and non-transitory computer readable medium for organizing scientific and technological publication data for performing predictive analysis is provided. According to embodiments of this disclosure, organizing scientific and technological publication data for performing predictive analysis includes: receiving the publication data from a publication database; parsing the publication data to obtain citations and associated citation data for each publication; storing the citation data in a first database; identifying one or more co-citations between publications in the citation data to obtain co-citation information between publications; and storing the co-citation information between publications in a second database.