143 results about "Source text" patented technology

The invention relates to the technical field of equipment maintenance and overhaul decision, and discloses a knowledge graph driven two-stage classification equipment condition-based overhaul decision method, which comprises the following steps: acquiring multi-source text data; performing first-stage classification on the equipment based on the equipment importance evaluation model, and performing second-stage classification on the parts based on the maintenance strategy; a device entity, a fault mode entity and a maintenance measure entity are taken as knowledge ontologies, state variable nodes are embedded, and a knowledge graph fusing state variables is constructed by using a long-short-term memory network; in combination with a semantic context sensing mechanism and a graph structure stability constraint mechanism, outputting candidate fault nodes; candidate maintenance measures corresponding to the candidate fault nodes are retrieved in the knowledge graph, candidate maintenance measure verification is carried out in combination with soft constraints and hard constraints, and a maintenance measure list is converted into an equipment maintenance decision scheme by utilizing an execution arrangement generator, so that intelligence of the maintenance decision scheme is realized in combination with the knowledge graph; and decision support is provided for maintenance personnel.

The invention relates to the technical field of building information, and provides a knowledge base construction and retrieval method and system based on a multi-source text in the building field, and the method comprises the following steps: a knowledge base construction stage: constructing a multi-dimensional metadata feature vector for a multivariate text based on a standard classification index table; the method comprises the following steps of: converting and segmenting a document, splicing an end clause with all superior title texts by utilizing a context inheritance algorithm to form a text unit with complete semantics, and performing dynamic filtering based on an analyzed query intention and a metadata vector at a user retrieval stage; then, in the screening set, performing fusion calculation on semantic vector similarity, keyword matching degree and authority offset weight based on effectiveness attribute and implementation time, and performing mixed retrieval and reordering on the text units; and finally selecting a text unit according to a sorting result and inputting the text unit into the large language model to generate answers. According to the method, high-precision and high-compliance intelligent retrieval and question answering of building domain knowledge are realized.

The invention relates to the technical field of prompt optimization engineering, in particular to an embedded clustering multi-path recall large model question and answer method, system and device and a medium, and the method comprises the following steps: preprocessing a multi-source text to generate an associated paragraph, and converting the associated paragraph into a semantic vector; executing clustering and adjusting a clustering center based on a matching strategy to form a vector library; receiving queries, converting the queries into semantic vectors and retrieving related units in parallel; the recall result is comprehensively evaluated on the basis of a credibility perception attention mechanism in combination with semantic matching and timeliness score, the semantic matching score is calculated by a local bge-ryanker-v2-m3 model, the timeliness score is quantized through a date difference attenuation function, and finally high-quality content is weighted and screened to be input into a large language model to generate question and answer output. And the result accuracy, timeliness and service adaptability are improved.

The invention discloses a structured data storage method based on natural language transformation, which comprises the following steps of: a system initialization configuration stage: deploying a protocol adapter in a local memory of a PC (Personal Computer) client, and loading natural language processing pipeline configuration parameters; a heterogeneous data acquisition stage: capturing a multi-source text data stream through the protocol adapter, uniformly converting the multi-source text data stream into a standardized data packet, and sending the standardized data packet to a message queue theme; a text cleaning stage: a named entity recognition stage: inputting the pure text data into an NER module deployed with a language model loader; in the conditional feature extraction stage, feature vectors are generated for texts meeting preset conditions on the basis of entity type tags in the entity recognition result; and a consistent storage stage: inserting the entity identification results into a relational database in batches, and updating the entity mapping relationship in the cache. According to the method, the intelligent level of cache management is remarkably improved, and the access fluency of the key data of the user is guaranteed.

In accordance with the described techniques, a system receives a plurality of facets describing language-agnostic aspects of language translation, a translation document describing language-specific rules for translating from a source language to a target language, and a source text in the source language. Using one or more machine learning models, a plurality of guidelines are extracted from the translation document and assigned to respective facets of the plurality of facets. The system translates the source text to a translated text in the target language using one or more machine learning models conditioned on the plurality of guidelines assigned to the respective facets.

A computer system generates descriptions of image-text misalignments. The system includes one or more processors and models for generating textual and visual descriptions of misalignments between a source text string and a source image. The textual description identifies misaligned text segments, while the visual description may include bounding boxes indicating the location of the misalignment. This system automatically generates synthetic image-text misalignment training examples and feedback, which includes generating misalignment captions and visual bounding box labels.

The invention discloses a text translation quality assessment method and device, computing equipment, a storage medium and a computer program product. The method comprises the steps of obtaining a translation pair formed by a to-be-assessed source text and a translation text; executing first-layer evaluation: evaluating the translation pair based on rule detection to obtain an evaluation result; executing second-layer evaluation, including evaluating the translation pair based on a thinking chain reasoning mechanism of a large language model to obtain an evaluation result; and executing fusion processing, including fusing the obtained evaluation results to obtain a target evaluation result representing the evaluation quality of the translation pair. Therefore, the translation quality evaluation method and device achieve the balance of high efficiency, high accuracy and interpretability of translation quality evaluation, and are particularly suitable for professional scenes such as game texts.

The invention discloses a similarity calculation method based on semantic editing fusion. The method comprises the following steps: calculating global semantic similarity of a source text and a target text by utilizing a semantic coding model; obtaining a first keyword list of the source text and a second keyword list of the target text through word segmentation processing, taking each first keyword in the first keyword list as a target word, and respectively forming a plurality of to-be-compared word pairs with a corresponding word in the second keyword list and a plurality of adjacent second keywords; based on the to-be-compared word pairs, calculating the semantic similarity between each group of target words and the corresponding words by utilizing a semantic coding model, and determining the editing distance between the source text and the target text; and determining the local semantic similarity of the source text and the target text according to the editing distance, and combining the global semantic similarity to obtain the final text matching similarity. According to the method, the problem that text matching only depends on character-level surface matching and neglects semantic association between word pairs is solved, and the text matching precision is improved.

The invention provides a method for generating high-quality instruction data in the ship manufacturing field based on a knowledge graph. The method comprises the following steps: (1) collecting and preprocessing a multi-source text; (2) knowledge extraction and graph construction; (3) knowledge point understanding evaluation and marking; (4) extracting constrained sub-graphs; (5) generating a multi-type instruction-response pair; (6) data source monitoring and map increment updating; (7) the instruction-response pair increment is updated and audited; and (8) performing multi-dimensional quality evaluation and output on the instruction data. The invention aims to solve the problems of high training data construction cost, high illusion rate, low coverage and poor timeliness in a large language model fine tuning process in the ship manufacturing field.

The invention discloses a user demand insight positioning method and system based on AI driving, relates to the field of artificial intelligence and big data analysis, and solves the technical problem that it is difficult to accurately recognize user demands and respond to the user demands in real time through dynamic analysis of multi-source text data and construction of a knowledge graph. Comprising the steps that text data are collected, structured data are generated through preprocessing, and a structured data unit is generated; calculating text vocabulary diversity and syntactic complexity indexes, inputting the text vocabulary diversity and syntactic complexity indexes into a pre-training regression model to obtain a dynamic character threshold value, encoding short / long texts by adopting a lightweight graph neural network and a hierarchical Transform, and extracting core demand elements; a dynamic time sequence knowledge graph is constructed, a time sequence edge weight updating algorithm is introduced to adjust node association strength, and cross-domain association is mined; and calculating a user demand intensity evaluation value based on an entropy weight method in combination with the inventory pressure factor and the group influence factor, thereby realizing inventory scheduling and personalized recommendation.

Provided in embodiments of the present disclosure are a content presentation method and apparatus, a device, and a storage medium. The method comprises: in response to a request for source text content, on the basis of analysis for the source text content, determining a first analysis result at least indicating summary information in the source text content and at least one second analysis result indicating at least one content unit of the source text content; acquiring respective role configurations of a plurality of speakers corresponding to the source text content; and generating first dialog content of the plurality of speakers on the basis of the first analysis result, the at least one second analysis result, and the role configurations of the plurality of speakers, wherein the first dialog content comprises a first dialog segment and at least one second dialog segment following the first dialog segment, the first dialog segment is associated with the summary information, and the at least one second dialog segment is separately associated with the at least one content unit.

This invention relates to the field of knowledge base construction technology, specifically disclosing a method and system for constructing an industry knowledge base based on text-based data synthesis. The method includes entity recognition of multi-source text data, constructing a knowledge point sequence, and marking key knowledge points in the knowledge point sequence; expanding the key knowledge points based on a large language model to generate an expanded knowledge point set; constructing a knowledge graph based on the expanded knowledge point set, and constructing a preset number of knowledge connection paths in the knowledge graph; classifying the knowledge connection paths and inserting them as knowledge skeletons into the industry knowledge base. When dealing with massive amounts of multi-source text data, this invention expands the keywords based on a large language model to obtain expanded knowledge points, then constructs a knowledge graph, extracts node paths from the knowledge graph, uses the node paths as knowledge skeletons, classifies them, and stores them in the knowledge base. While the resulting knowledge base still contains a large amount of information, it significantly simplifies its size.

The application discloses a kind of general pre-training graph structure large model construction method and system, belong to graph computing technical field, including: from the sample set of text attribute subgraph and node text attribute contained in multiple source text attribute graph data is constructed;Graph structure encoder is pre-trained using the reversible serialization mode based on Euler path improvement mode, and graph structure representation is obtained;Text feature encoder is used to encode node text attribute into text semantic representation, after graph structure representation and text semantic representation are mapped to the same alignment space by mapping network, based on text semantic representation and graph structure representation, construct joint global representation alignment task, substructure-phrase level local semantic alignment task, and graph to text reconstruction task, train mapping network, obtain general pre-training graph structure large model, so it can effectively bridge the modal gap between graph structure and text semantic, significantly improve the generalization ability and reasoning reliability of model in downstream task.

The present disclosure provides a method for knowledge distillation, an apparatus, an electronic device, a medium and a computer program product. The method includes: acquiring a training source text and a standard translation text corresponding to the training source text; inputting the training source text into a teacher translation model and a student translation model separately, to obtain a teacher distribution output by the teacher translation model and a student distribution output by the student translation model; obtaining a standard translation distribution according to the standard translation text and the training source text; and performing iterative training on the student translation model according to the teacher distribution, the student distribution, and the standard translation distribution, to obtain a target machine translation model.

A system and method for transforming text within documents using, such as by using large language models (LLMs). Users can select source text from a source document, in response to which a painting configuration is identified or generated based on the source text, such as by providing the source text and a source prompt to a large language model to produce source output, and selecting or generating the painting configuration based on the source output. The user can select destination text, in response to which the painting configuration is applied to the destination text, such as by selecting or generating a destination action definition based on the painting configuration and the destination text, and providing the destination action definition to a large language model to produce destination output. The destination text may be replaced with the destination output, or output derived therefrom. In this way, the system can extract a variety of sophisticated properties, such as style or tone, from user-selected text source text, and apply those properties to user-selected destination text, with minimal user input.

The invention discloses a gas hidden danger investigation semantic element extraction method, which comprises the following steps: inputting a gas hidden danger investigation field knowledge text, and carrying out hidden danger knowledge segment segmentation on the knowledge text; constructing a candidate hidden danger semantic element set for the segmented hidden danger knowledge segments; calculating semantic element space risk measurement based on the hidden danger semantic element set to obtain a risk score; and recognizing the hidden danger semantic elements based on the risk scores. According to the method, the accuracy of gas hidden danger recognition can be improved, the troubleshooting accuracy, efficiency and risk disposal priority are remarkably improved, and the method is suitable for knowledge extraction and hidden danger recognition scenes of various multi-source texts.

The application discloses a cross-domain named entity recognition method based on an autoregressive model, and comprises the following steps: S1, encoding an input sequence; S2, encoding a label through a label encoder; S3, obtaining label background information; S4, obtaining label context information; S5, connecting the label background information to the input sequence and connecting the label context information to the predicted named entity label as final label perception information z i , and finally obtaining a final sequence representation u; the application provides a cross-domain named entity recognition method based on an autoregressive model, which improves the relationship between a source text and its named entity label, improves the portability of label information, and helps the model to promote domain adaptation.

The invention discloses a two-stage intelligent text clustering method based on large language model guidance, and relates to the technical field of artificial intelligence, natural language processing and machine learning, and the method comprises the steps: converting a to-be-processed text into a multi-source text vector set, and determining a standardized unified text representation vector set in combination with information entropy; determining a pre-clustering result through the dynamically calculated initial clustering number; calculating an inter-cluster Ward distance based on a pre-clustering result, introducing cluster size weighted correction, and constructing a clustering tree; a candidate cluster number set is determined based on the MDL principle, cluster pairs to be combined are extracted by accessing a cluster tree in a reverse order, whether the sample pairs belong to the same cluster or not is judged through a large language model, and a weighted F-beta score is calculated to determine the optimal cluster number; and intercepting a target clustering result according to the optimal clustering number and generating a clustering result. According to the method, the calculation complexity is reduced to O (n), the consistency of a clustering result and human cognition is improved, and the method is suitable for rapid and accurate clustering of large-scale text data.

This invention discloses a sentence-level question generation method based on syntactic-aware prompt learning. First, a bidirectional syntactic dependency graph is constructed based on a given sentence. Its semantic representation is obtained through a relation-aware attention graph encoder. The encoded vectors are then input into a softmax layer, and the top k vectors are selected as continuous prompts based on probability. The prompts are concatenated to the given source text and the answer using prefix adjustment, and then input into a BERT model for encoding. The encoded result is then fed into a Transformer model for decoding. At each time step of decoding, the syntactic dependency information of the generated text sequence is modeled. This information, combined with the syntactic dependency information of the source sentence, determines the parts that the decoder needs to focus on, assisting in the generation of the current word. Simultaneously, a copying mechanism is introduced to handle situations where the generated word is not in the question vocabulary, allowing the model to directly copy words from the source text.

[Problem] To compare information disseminated in an unspecified field with information extracted from other information sources. [Solution] A program for operating a computer causes a processor of the computer to execute: a step for receiving designation of a text; a step for generating a prompt including an instruction to extract, from an information source usable by an artificial intelligence system, at least one source text including a description similar to a designated text which is the text that has been designated, and to output information on the source text, and an instruction to compare the extracted source text with the designated text that has been designated; and a step for, on the basis of an answer obtained from a large-scale language model by inputting the prompt to the large-scale language model provided by the artificial intelligence system, presenting, to a user, information on the source text and a result of the comparison between the designated text and the source text.

The embodiments of this disclosure relate to post-editing methods and apparatuses, electronic devices, and storage media, and pertain to the field of machine translation technology. The post-editing method includes: acquiring a target source text and a target machine-translated text, wherein the target machine-translated text is the machine-translated version of the target source text; inputting the target source text and the target machine-translated text into a pre-trained post-editing model, and correcting the target machine-translated text using the post-editing model; the correction methods of the post-editing model include deleting tokens, inserting placeholders, and replacing placeholders with tokens; and outputting the target post-edited text obtained by the post-editing model correcting the target machine-translated text.

The application discloses a content-tracing-based enterprise culture hot word analysis method, and relates to the technical field of enterprise culture analysis.The method comprises the following steps: collecting culture-related original data from three dimensions, and constructing an enterprise-specific culture term library through preprocessing, density peak clustering, de-redundancy and matching with top-level culture elements; classifying enterprise internal texts to be analyzed according to departments and text types, automatically matching terms of corresponding categories in the enterprise-specific culture term library, and performing directional word segmentation based on preset rules; calculating the scene coverage of terms according to department scenarios and text type scenarios, calculating the culture correlation degree by using Jaccard similarity, and screening core hot words by combining the Apriori algorithm; locking target hot word origin texts based on the core hot words, counting the track data of the origin text transmission channels, identifying the transmission nodes and constructing the transmission track; and the method improves the accuracy of hot word analysis and provides support for targeted promotion of enterprise culture.

The application discloses a knowledge base and multi-agent collaborative academic review automatic generation method, system, device and storage medium, relates to the field of artificial intelligence and natural language processing technology, and comprises the following steps: adopting a mixed extraction mechanism to extract an agent to original unstructured literature, and constructing a bottom-layer JSON knowledge base containing technical indexes and quantitative data; mapping the knowledge base information to a preset target hierarchical structure through a planning agent, and generating a global review outline; adopting a block retrieval enhancement strategy, a writing agent carries out chapter text synthesis and graph generation under the constraint of forced binding quantitative evidence; comparing the source text and the generated text through entity lexical analysis, and performing quality evaluation.The application adopts the above-mentioned knowledge base and multi-agent collaborative academic review automatic generation method, system, device and storage medium, effectively improves the local fact accuracy and global logical coherence of the generated literature, and realizes automatic and high-fidelity compilation of the academic review.

The invention relates to the technical field of retrieval enhancement generation, and provides a retrieval enhancement method based on domain-term association mining and term closure expansion, and the method comprises the steps: obtaining an original term set covering a professional domain from a professional domain resource text; obtaining a terminology set; obtaining a terminology set screened by the reflection mechanism; obtaining a final terminology set; generating a plurality of term pairs; the screened term pairs are obtained; obtaining a term relation graph; receiving a natural language question input by a user, and constructing a term set corresponding to the natural language question; obtaining a term extension closure set; generating a candidate paragraph set; inputting the natural language question and the candidate paragraph set into a semantic reordering model, outputting a correlation score of each paragraph in the candidate paragraph set by the semantic reordering model, and taking the first K paragraphs with the highest correlation scores as a final support content set; and the large language model outputs an answer text corresponding to the natural language question.

The invention relates to the technical field of intelligent data checking, in particular to a digital file data checking method and system.The method comprises the steps that high-frequency words under all themes in each time of division and high-frequency words under any theme in other time of division are obtained based on the overlapping degree between the high-frequency words under all themes in each time of division; determining the importance degree according to the high-frequency words and keywords of the business file text under each theme and the overlapping degree of the high-frequency words and keywords of the business file text under each theme; based on the overlapping degree between all high-frequency words of the business file text under each theme and all keywords of the to-be-checked file text under each division, and in combination with the importance degree, all high-frequency words and the probabilities of all high-frequency words belonging to the themes, determining theme weights and feature vectors so as to determine semantic vectors of the business file text; and matching the to-be-checked file text. According to the method, the problem of low efficiency caused by the fact that deep semantics cannot be understood when a multi-source text is processed by a traditional data checking method is solved, and the data checking efficiency is improved.

The embodiment of the invention provides a log analysis and risk analysis method and system. The method comprises the following steps: pre-compiling a rule configuration file to generate a regular object, and storing the regular object in a pre-compiled rule list; wherein the regular object is packaged with a plurality of fields; performing log format adaptive analysis on the to-be-analyzed data source text to generate a log message; cyclically traversing the rule list, searching content matched with the regular object in the to-be-analyzed data source text so as to filter repeated results and determine to generate alarm description; constructing the log message and the alarm description into a de-duplicated data pair; performing cache check operation on the data pair to generate a first cache result; performing prediction operation on the data pair batch calling AI model to generate a second cache result; and generating a prediction result based on the first cache result and the second cache result. The problem that the efficiency of log analysis and risk analysis work is low in the prior art is solved.

This application discloses a training method, apparatus, computer device, medium, and program product for a general translation model. Semantic tags are added to language sample pairs used for training. The semantics represented by these tags enable the general translation model to clearly define the translation task it performs. For example, based on the target semantic tags and the source text, an initial general translation model is used to translate the source text. The initial general translation model translates the source text based on the translation task represented by the target semantic tags, resulting in translated text. Based on the differences between the translated text and the target text, the model parameters of the initial general translation model are adjusted to obtain the general translation model. The general translation model can not only learn how to translate text belonging to the source language into text belonging to the target language, but also learn to understand the semantics of the semantic tags, thereby clarifying the target language indicated by the current translation task, reducing the probability of off-target problems during translation, and improving translation accuracy.

An inference acceleration method relating to artificial intelligence technical fields such as a large model, deep learning, and natural language processing is provided. The inference acceleration method for large models includes: after inputting a source text to be processed into a target large model, obtaining a top-layer hidden state of the target large model for predicting a next token; obtaining action decision information corresponding to the next token according to the top-layer hidden state; in response to determining that the action decision information is a copy action, obtaining a text copy interval corresponding to the next token according to the top-layer hidden state; copying text in the source text to be processed that is located within the text copy interval, and using a copy result as the next token.

The application discloses an industrial document translation method and system based on visual positioning, and belongs to the field of document translation. The application comprises the following steps: S1, performing geometric correction on an input industrial document image, and identifying the document type of the corrected industrial document image; S2, extracting a source text area from the identified document type, translating the extracted source text area through a hybrid search engine, generating a corresponding translation text, and performing structure consistency verification; S3, selecting a placement strategy of the translation text box based on the page geometric information of the industrial document and the translation text, scoring each placement strategy of the translation text box, and selecting an optimal placement strategy according to the score; and S4, rendering the translation text box on the industrial document according to the selected optimal placement strategy, and establishing visual association between the translation text and the source text, thereby improving the spatial layout adaptability of the industrial document translation, and solving the problem of invalid spatial layout adaptability of the industrial document translation in the prior art.

The application discloses an electromagnetic space field entity relationship joint extraction method with additional time information, comprising the following steps: obtaining social platform open source text data by using a Python crawler technology; performing text cleaning and preprocessing on the obtained open source text to form a data set C; determining entity and relationship classification, and formulating an entity relationship sequence joint labeling strategy; labeling the data set C with a triple tag; the triple tag is (main entity Subject, relationship Predicate, and object entity Object); an entity relationship joint extraction model STERM based on fragment sorting is constructed, and an entity relationship triple prediction model is obtained through training; and a (main entity, relationship, object entity, and time) quadruple list is obtained through a sequential priority nearest matching algorithm. The method solves the problems of error accumulation and exposure deviation, entity nesting and relationship overlap, and the problem that an entity relationship triple cannot represent dynamic changes of a relationship.