36 results about "Tag system" patented technology

Techniques for constraining a language model's generation / decoding process using a finite state machine are described. A finite state machine may represent information corresponding to APIs, arguments and argument values that are available / supported by the system. In some embodiments, a language model (LM) is constrained to generate tokens representing valid API calls based on the finite state machine. The system may enable generation of argument values from a defined set or free-form generation of argument values. The system may also enable unconstrained generation of a response by the LM. In some embodiments, a trie data structure is used to determine the possible next tokens that the LM can generate from.

The invention discloses a mass data automatic label generation method based on a large model and a rule engine, which comprises the following steps: S1, multi-source data acquisition and preprocessing: acquiring data through a unified data interface, and then performing data cleaning, data standardization processing and data format conversion to obtain standard input data; s2, constructing a label system framework; s3, generating a large model label; s4, performing intelligent clustering and label refining; s5, rule engine constraint and optimization; s6, evaluating and optimizing label quality; and S7, automatically expanding and updating the tag system. According to the method, automatic label generation of mass multi-source heterogeneous data is realized through a large model and rule engine technology under the condition of no personnel intervention, a large-scale label system containing thousands of labels is constructed, meanwhile, the label system can be rapidly updated along with product iteration and keep timeliness, and the generated labels are accurate and extensible.

Systems and methods for detecting anomalies in generative outputs are disclosed herein. The system receives a user prompt indicating a request for data over a time period. The system inputs, into a model, the user prompt to cause the model to generate an output based on the user prompt. The system then generates the first tokens based on the output. To generate the second tokens, the system retrieves, based on the user prompt, sources relating to the data requested by the user prompt. The system then generates queries to request, from the sources, the data over the time period and generates the second tokens based on the retrieved data. The system then performs a comparison of the first tokens and the second tokens and accepts or rejects the output of the model based on the comparison.

The invention relates to the technical field of property right project management, in particular to a dynamic management system and method for property right project state tags. The method comprises the following steps: a data acquisition and processing unit acquires project data and preprocesses the project data; a state label system construction unit constructs a multi-level state label system by using a property right project full life cycle process based on the preprocessed project data; and the state reasoning and dynamic updating unit is used for identifying the change trend of the project state based on the multi-level state tag system by adopting a level state machine model, generating a state tag set, dynamically updating the tag set when a project business event occurs, and generating a dynamically evolved state tag set. According to the method, a hierarchical state machine model and a link correction mechanism are introduced, a label set is dynamically updated when a service event is triggered, and structure-level compensation is carried out on inference interruption caused by data missing or cross-system inconsistency.

The invention provides a multi-label path reasoning method and system, a medium and a terminal. The method comprises the following steps: constructing a label system; dynamically updating a keyword set in the tag system; and based on the tag system and the keyword of the tag node, allocating a final tag to the to-be-classified text. According to the multi-label path reasoning method and system, the medium and the terminal, multi-label attribution can be efficiently and accurately completed under a complex label system through label system structured grouping, upper and lower layer path collaborative judgment and an embedded semantic fusion and path scoring mechanism.

The invention belongs to the technical field of spacecraft product quality and reliability analysis modeling, and particularly relates to a fault tree generation method and device based on a knowledge graph and a large model fusion technology. The method comprises the following steps: constructing a spacecraft product fault domain ontology model, establishing a tag system comprising entity tags and relation tags, and extracting related knowledge from a technical document, a SysML model and an FMEA table by using a large language model to form a triple; according to entity types and relation types of different source knowledge, fusion is carried out, and a fault knowledge graph is constructed; obtaining a function dependence sub-graph, a fault behavior sub-graph and a redundant sub-graph of a function or a composition structure from the fault knowledge graph, and retrieving a specific sub-graph from the fault knowledge graph according to a user question; analyzing fault propagation relation and redundant structure information based on function dependence in the specific sub-graph to construct an OR gate and an AND gate, and mapping nodes in the specific sub-graph to a basic event, an intermediate event, a top event and a gate to generate a fault tree.

The invention discloses a germplasm resource electronic tag system which comprises an electronic tag which adopts an RFID tag or a two-dimensional code tag and stores a unique identification code of a coded germplasm resource. According to the method, the germplasm resource management efficiency is improved; through combination of the electronic tag and the data management platform, rapid acquisition and updating of germplasm resource information are realized; a worker does not need to manually look up a large number of paper records or switch query in a plurality of systems, and can instantly obtain comprehensive germplasm resource information only by scanning the electronic tag through the handheld terminal, so that the time and the labor cost are greatly saved; for example, in germplasm resource warehouse-in and warehouse-out management, detailed information of each batch of germplasm resources needs to be manually recorded in a traditional mode, time consumption is long, and errors are prone to occurring; when the system is used, a worker scans the electronic tag by using the handheld terminal, the recording and updating of the warehouse-in and warehouse-out information can be quickly completed, the whole process only needs several seconds, and the working efficiency is greatly improved.

A system for generating a labeled dataset is provided. The system includes a processing component configured to: receive first data, wherein the first data comprises one or more frames, and wherein the first data comprises data defining an object of interest (203) within a predetermined area (202); receive second data, wherein the second data is associated with the object of interest within the predetermined area; analyze the one or more frames of the first data to identify the object of interest present in the first data based on the second data; label the one or more frames of the first data based on the analysis to generate a labeled dataset; and output the labeled dataset. A method for generating a labeled dataset, a system for training a machine learning model, and a detection system for detecting one or more objects of interest are also provided.

This invention relates to the field of data parsing technology, specifically a method, system, and medium for parsing non-standard nested format data with zero code. First, sample data is prepared, obtaining non-standard format samples to be parsed. A unique marker is generated for each character in the interactive text area through a rendering module. Then, single-level parsing rules are generated, identifying the content field type marked by the user in the interactive text area and generating parsing rules. Next, nested parsing rules are configured, including a tree-like nesting strategy. Finally, production data is parsed, and the parsing rules are distributed to the data access module for batch parsing of the input production data. This allows users to simply mark the sample data visually, and the system automatically captures mouse events, eliminating the need for users to write or understand regular expressions. Simultaneously, sample fragments are instantly converted into executable rules, shielding underlying syntax and conversion details, achieving zero code and zero syntax requirements. Parsing rule definition can be completed independently through interactive word highlighting.

The invention relates to the technical field of electronic commerce and information recommendation, in particular to a recommendation method and system based on user tags. According to the method, the importance of the user tags can be effectively determined by constructing the user hierarchical tag system and calculating the weight of the ungenerated multi-dimensional user tag set; periodically attenuating and updating the weights of the tags in the user tag set by using a time attenuation function, and capturing interest drift through dynamic updating; a commodity feature vector calculated by a commodity-label knowledge graph is constructed based on commodity data, so that commodity features can be effectively determined; the similarity between the commodity feature vector and the user feature vector is calculated, and the recommended commodity is determined based on the similarity, so that the recommended commodity can be accurately determined; and determining a recommendation state based on a comprehensive score determined by the click rate, the conversion rate and the stay duration of the user on the commodity recommendation page in a preset period, and adjusting corresponding parameters based on the recommendation state. The recommendation efficiency is improved.

The application discloses a member right management system and method based on data tags, and relates to the technical field of big data analysis.The application forms a structured analysis set by associating scattered member attributes and right use data according to unique identification by constructing a three-dimensional member tag system, and realizes fine classification of member groups.Based on standardization processing and quantitative analysis, the application calculates the right use heat and the correlation strength characteristics by weighting, constructs a scientific adaptive scoring model by combining Pearson correlation analysis, accurately describes the group right preference and core attributes, realizes accurate screening of target member sets, and improves the pertinence and rationality of right allocation.The application sequentially issues rights through a phased verification promotion mechanism and dynamically monitors abnormalities, simultaneously constructs a multi-dimensional effect evaluation system to form a dynamic closed loop, ensures the safe and stable updating of rights, responds to changes in member needs in real time, breaks the static and solidified configuration mode, and realizes the intelligentization and sustainable iteration of right management.

The application relates to a label system-based Internet of Things monitoring model construction method and belongs to the field of natural language processing. The method comprises the following steps: establishing an initial label system, manually labeling a structured data set based on the initial label system, and constructing a first training sample set; constructing a Bert model, training the Bert model based on the first training sample set to obtain a trained Bert model; inputting business data into the trained Bert model to obtain corresponding characteristic values, and obtaining a perfect label system based on the characteristic values; constructing an Internet of Things monitoring model, using the Internet of Things monitoring model to predict business data to obtain predicted business data; using the perfect label system to label the predicted business data to construct a second training sample set; and training the Internet of Things monitoring model based on the second training sample set to obtain a trained Internet of Things monitoring model. The method can automatically generate feature labels required by business, label business data by using a perfect label system, and improve the recognition effect and training efficiency of the model.

The invention provides a dynamic electronic tag system for railway communication equipment and an implementation method of the dynamic electronic tag system, relates to the technical field of intelligent operation and maintenance of the railway communication equipment, and solves the technical problems that a traditional paper tag is not easy to modify and information updating is lagged. The system comprises a background management platform used for storing and managing equipment basic information and port service information of railway communication equipment; the two-dimensional code generation module is used for generating a two-dimensional code image; the physical two-dimensional code label is used for being pasted on a panel of the corresponding railway communication equipment; the information association and synchronization module is used for updating the port service information of the background management platform in real time to obtain real-time port service information; and the mobile terminal is used for scanning the physical two-dimensional code label, sending the obtained two-dimensional code information to the background management platform for verification and decryption, and receiving and visually displaying the returned real-time port service information. The method is used in the intelligent operation and maintenance management process of the railway communication equipment.

The invention discloses a user tag system testing method and device, equipment, a storage medium and a program product. The method comprises the following steps: constructing and storing a label data set, and synchronizing the label data set to a big data label system; each piece of label data in the label data set takes a label identifier representing a label type and an enumeration value as indexes; simulating to obtain a first test case, and utilizing the first test case to obtain a first test result from the label data set; the first test case represents a circle selection rule of the user; based on a request sent to the big data label system, a second test result of the big data label system based on the request is obtained, and the request comprises first information related to the first test case; and comparing the first test result with the second test result, and determining whether the big data label system passes the user circle selection processing test according to the comparison result, thereby ensuring the accuracy of the big data label system on the user circle selection processing result.

The invention provides an automatic data annotation and quality verification method for a life service scene. The method comprises the following steps: a rule base definition step: constructing a label system for the life service field; converting the label system into a configurable rule base, wherein the rule base comprises dominant rules based on keyword matching and implicit rules based on semantic analysis; a machine labeling engine step: implementing a two-way parallel labeling architecture based on the rule base, wherein the two-way parallel labeling architecture comprises a dominant feature processing path and a semantic feature processing path; a quality verification and optimization step: carrying out confidence evaluation on machine labeling results, and dividing the results into a high-confidence automatic adoption group and a low-confidence manual verification group according to a preset threshold value; and correcting the low-confidence data to generate a weighted labeling sample, and realizing iterative optimization of a labeling model. The invention aims to solve the problems of the existing data annotation mode in the field of life service, and realizes efficient, accurate and consistent data annotation and quality guarantee.

The invention discloses an intelligent and accurate topic label labeling method based on a general large model, and relates to the technical field of data management, and the method mainly comprises the steps: constructing a structured label system as an output template through fusing a general basic label dimension and a subject exclusive label dimension; performing multi-modal preprocessing on the question, analyzing text, formula and graphic information, and fusing to generate standardized question representation; inputting the representation into a general large model subjected to subject field data fine adjustment, and through a cue word guiding model, analyzing the content of the question according to a tag system and generating a structured multi-dimensional tag; performing automatic logic verification and correction through a knowledge graph and a rule, and collecting, feeding back and constructing an optimization sample through a manual recheck interface; and performing incremental fine adjustment on the model based on an optimization sample, and performing versioning updating on the label system, thereby realizing closed-loop continuous optimization of the label system. According to the technical scheme of the invention, the intelligent, precise and continuous optimization of topic labeling is realized.

The invention discloses an equipment fault diagnosis dynamic optimization method and system based on an intelligent power plant, and relates to the technical field of intelligent operation and maintenance of the intelligent power plant. Standardized acquisition and feature construction of multi-source operation data are carried out through unified coding, time synchronization and a semantic tag system; a mechanism model and AI model dual-channel fusion diagnosis engine is constructed, and parallel reasoning and structured diagnosis of multi-source data are output through residual calculation and consistency analysis; a dynamic optimization model containing three elements of energy efficiency, time delay and risk is constructed based on structured diagnosis, and operation parameters are reconfigured in real time through constraint solution and a control quantity mapping mechanism. According to the method, multi-source data fusion diagnosis and dynamic optimization linkage is achieved, the real-time analysis, self-adaptive optimization and continuous learning capabilities are achieved, and the intelligent level of equipment operation is improved.

The application provides a form generation method and system for authenticating documents, and relates to the technical field of data processing. The method comprises the following steps: obtaining an authenticating document; converting the authenticating document into an XML file to map the data of the authenticating document to an XML tree node of the XML file; standardizing the XML file through a globally unique identifier and a tag system to obtain a standardized XML file; combining the standardized XML file to extract template candidates in the authenticating document that conform to different manageable data templates; combining the template candidates to map the authenticating document to a target form; and establishing a bidirectional mapping relationship between the target form and the authenticating document to realize bidirectional conversion between the target form and the authenticating document and complete the form processing of the authenticating document. Through the establishment of the bidirectional mapping relationship, the synchronous updating of the form and the document content can be realized, the error of manual operation is avoided, and the automation, accuracy and practicability of the document conversion are significantly improved.

The application provides a posture estimation method of a tag device, a space interaction system and an electronic device. The posture estimation method of the tag device comprises the following steps: determining a tag coordinate system according to the position of the tag device, and determining an accelerometer coordinate system according to the position of an accelerometer of the tag device. The gravitational acceleration of the tag system is determined. An anchor-tag rotation matrix set is determined. A target rotation matrix is determined according to the relative orientation angle between the tag device and an anchor device. The posture estimation information of the tag device in the anchor coordinate system is obtained according to the target rotation matrix. Therefore, the posture estimation method, the space interaction system and the electronic device provided by the application can reduce the design distance of the antenna on the tag device, reduce the antenna complexity of the tag device, and thus meet the light and thin, small and compact requirements of the tag device. Meanwhile, the posture estimation accuracy of the tag device on the anchor device side can also be improved.

This invention discloses a method and system for processing archival information based on digital security, relating to the field of digital data processing technology. The method includes the following steps: acquiring the first metadata of the archival information; based on a preset tag system template, using a hierarchical logic combining rule matching and manual review, identifying sensitive and non-sensitive tags in the first metadata; the preset tag system template dynamically iterates through historical sample learning and new tag case verification; for the first metadata corresponding to sensitive tags, differentiated desensitization processing is performed according to the sensitive tags; for the first metadata corresponding to non-sensitive tags, no desensitization processing is performed, and the data is integrated to obtain the second metadata. Through efficient collaboration across three dimensions and a dynamically iterative tag system, the accuracy of identification is improved; hierarchical identification improves efficiency, ensures quality, and reduces costs; differentiated desensitization processing protects security and reduces losses; and integrated metadata facilitates the secure and efficient circulation of archives.

This invention relates to an automatic advertising video generation method and system based on multimodal input, comprising: receiving multimodal input of text instructions and product images; using an advertising tag library matching module, extracting product image features based on a preset tag system constructed from advertising samples, and performing similarity matching with the tag library to select the best advertising sample; calling a large language model, using the best advertising sample as a reference, to generate a complete storyboard script including scene setup, product feature display, and core advertising slogan; using a prompt word engineering module, integrating storyboard text descriptions, product reference image visual features, and preset advertising duration to generate video generation prompt words; outputting the advertising video based on the prompt words, while simultaneously optimizing the tag library and matching algorithm through manual annotation. This solves the problems of cumbersome traditional advertising generation processes and the disconnect between creativity and visuals, achieving efficient and accurate automatic generation of advertising videos, and can be widely applied to various product advertising production scenarios.

This invention discloses a streaming tag computation method based on event-driven and metadata-driven approaches, addressing issues such as poor real-time performance in IPTV tag management, high rule modification costs, untraceable status, and difficulties in tag system expansion. This invention constructs a real-time data stream access platform and a tag rule metadata repository, decoupling rules from computation tasks; it generates real-time computation tasks based on metadata-driven approaches and completes incremental tag computation through event-driven mechanisms; it supports dynamic hot updates of rules, establishes a closed-loop management system for the entire tag lifecycle, and designs a unified data model and hybrid storage architecture for multiple tag types. This invention pioneers a collaborative mechanism between event-driven streaming incremental computation and metadata-driven hot updates, achieving tag synchronization latency ≤1 second, a 90% improvement in rule modification efficiency, and 100% traceability of status transitions, providing IPTV operations with real-time, flexible, and controllable tag capabilities.

The invention discloses a label extraction method and a related device. The method comprises the steps that a to-be-processed statement is analyzed by utilizing a target model in combination with a multi-level label pool, at least one multi-level candidate label corresponding to the to-be-processed statement is obtained, the multi-level label pool comprises at least two levels of label systems, and different levels of labels correspond to semantic classifications of different granularities; and screening the at least one multi-level candidate tag based on a mutual exclusion relationship among different tags in the multi-level tag pool to obtain at least one target multi-level feature tag corresponding to the to-be-processed statement, so that more refined semantic capture is realized through at least two levels of tag systems in the multi-level tag pool; the problem of information omission or semantic deviation possibly caused by a single-granularity label is avoided, semantically incompatible label combinations are excluded by utilizing a mutual exclusion relationship among different labels in the multi-level label pool, target model understanding deviation caused by user semantic fuzziness or errors is reduced, and the accuracy and robustness of label extraction are improved.

The application discloses a kind of based on label system's internet of things service processing method and device.The method comprises the following steps of: placing multiple label identifiers in extensible tree structure to form label system;Binding is carried out to all data objects in internet of things platform and label identifier, and label dictionary is formed, and label index table is established;Based on the label index table, the service request of the natural language form received is converted into the service request of label form, the service request of label form is executed according to the logical service rule corresponding to the label, and the processing result of the service request is generated.The technical scheme of the present application realizes full object labeling by constructing tree-shaped extensible label system, effectively releases the strong coupling of underlying data and business.Supports no-code business expansion, so that the platform can be flexibly adapted to multiple scene requirements, while greatly reducing development and maintenance cost, enhance the compatibility and running stability of system.

The invention relates to the technical field of digital reading platforms, in particular to a retrieval question and answer system and method based on multi-source hierarchical context, and the system comprises a background service module, a book label management module, a session reading management module, an RAG question and answer processing module and a member and Token fine control module. The book label management module, the session reading management module, the RAG question and answer processing module and the member and Token fine control module are all connected with the background service module. The problems that in the prior art, context sensing fineness is insufficient, multi-source content collaboration is difficult, a label system is scattered, and system expansibility is weak are effectively solved, the question and answer accuracy and scene adaptability are remarkably improved, meanwhile, refined management and control of resources and permissions are achieved, and the user experience is improved. And an efficient, extensible and coherent-experience integrated solution is provided for knowledge services.

The invention relates to the technical field of database structures, in particular to a dynamic optimization system and method based on label nebula density analysis for a large-scale business system and a storage medium. The optimization system comprises a data connection and acquisition module; a label star map construction and calculation module; a density analysis module; an optimization suggestion generation module; according to the method, an abstract tag quality problem is converted into a concrete graph network density problem, and quantitative and visual diagnosis of the health degree of a tag system is realized. The method not only solves the problem of'overfitting ', but also solves the problem of'missing', and meanwhile, can reversely promote the improvement of the content of a service system.

The disclosed methods include: detecting data associated with a first tag comprised in the tag network; resolving the data to generate resolved data associated with a plurality of tags in the tag network including the first tag and a second tag; determining the second tag based on the resolved data; and extracting or determining, using the resolved data: first telemetric data associated with the first tag, first location data associated with the first tag, second telemetric data associated with the second tag, and second location data associated with the second tag. The methods also include formatting the first telemetric data, the first location data, the second telemetric data, and the second location data to generate an inventory map associated with the first mobile or stationary cargo and the second mobile or stationary cargo; and transmitting the inventory map to a display device configured to visualize the inventory map.

Systems and methods are provided for automated metadata tag score determination that leverages machine learning (ML) models and weighted knowledge graphs to obtain and assign metadata tag scores to instances of underlying data. Examples include generating a domain-specific knowledge graph related to a particular domain. The domain-specific knowledge graph comprising a plurality of domain-specific features connected via a plurality of domain-specific relationships. Responsive to receiving an input tag, examples extract domain-specific features of the plurality of domain-specific features associated with the input tag and a subset of domain-specific relationships of the plurality of domain-specific relationships corresponding to the subset of domain-specific features. The examples then determine metadata tag score corresponding to the input tag for items of the particular domain based on the subset of domain-specific features and the subset of domain-specific relationships and update metadata for each item to include the determined metadata tag scores.

A method comprises generating a ciphertext rule defining a column tag when at least a subset of digits in a ciphertext stored in a column of a table matches predefined ciphertext digits associated with at least one of a first purpose, a first type, or a first sensitivity of column data in the column, generating, by the rule application, a table rule defining a table tag when the table includes the column that is tagged with the column tag, wherein the table tag labels the table as being associated with at least one of a second purpose, a second type, or a second sensitivity of table data in the table, and tagging, by a tag application executing at the database management system, the column with the column tag based on the ciphertext rule and the table with the table tag based on the table rule.

The embodiment of the invention discloses a service management method and device based on a smart tag, equipment and a medium. The method comprises the following steps: acquiring multi-data-source integrated data from a plurality of different types of business systems in real time; constructing a first map based on a preset map generation rule; based on a preset label system, generating node labels corresponding to the nodes in the first atlas to obtain a second atlas; if the special data conforming to the preset business event exists, adjusting a special label in the special node based on a special label generation rule corresponding to the special data to obtain a third graph; performing hidden label mining processing based on a preset self-learning label generation rule, and associating the obtained hidden label to a node corresponding to the third map to obtain a fourth map; and according to the label of each node in the fourth map and a preset label service action mapping rule base, generating and executing a target service action instruction. By implementing the method provided by the embodiment of the invention, the business management efficiency can be improved.