74 results about "Semantic code" patented technology

The invention provides a multi-round dialogue intention recognition method and system based on self-adaptive semantic understanding, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a natural language dialogue text of a current round of a user, and taking the natural language dialogue text as original input data; based on original input data, multi-level semantic features are extracted through a dynamic semantic coding algorithm, and semantic vector representation of a current round of dialogue is generated; setting three fixed anchor points in a semantic vector space based on a current round semantic vector and a historical dialogue state vector to form a triangular analysis structure; performing gridding segmentation on the triangular analysis structure, and generating a feature adjustment value according to distribution characteristics of segmented grids; and dynamically correcting the extraction process of the context-related features by using the feature adjustment value to obtain the corrected context-related features. According to the method, end-to-end optimization is realized in multiple rounds of interaction scenes such as customer service and intelligent assistants through full-process design.

The invention relates to the technical field of artificial intelligence and language learning, in particular to a multi-modal language learning auxiliary system and method based on artificial intelligence, and the system comprises a multi-modal input module, a cross-modal feature fusion module, a dynamic adaptive learning module and an interactive feedback generation module. Wherein the multi-mode input module is used for receiving original text data, original image / video data and original voice data; the cross-modal feature fusion module is used for extracting a visual feature vector and a semantic coding vector and generating a cross-modal joint feature vector; the dynamic adaptive learning module is used for generating dynamic scene parameters; and the interactive feedback generation module is used for outputting a multi-mode feedback data packet through a prompt learning engine. According to the method, by fusing multi-modal feature alignment and dynamic context parameter modeling, linkage generation of grammar, culture and pronunciation feedback is achieved, and the context understanding ability and interaction feedback precision in the language learning process are improved.

The invention provides a multi-modal database query engine and method based on unified semantic representation, the engine associates and stores a vector library of unified semantic vectors of entities, and the core comprises five modules: a query analysis and semantic coding module for processing single-modal and multi-modal queries; analyzing the semantic structure and generating a corresponding semantic vector or a fused semantic vector; the query execution plan generation module generates an optimal plan including a retrieval strategy, weight allocation and resource scheduling according to semantic features, vector library features and the like; a multi-modal query execution module queries a vector library according to a plan to obtain candidate results; the query result fusion module performs weighted fusion on the candidate results to generate weighted scores of the candidate results; and the output module outputs a result according to weighted score sorting. The engine realizes efficient processing of multi-modal query and accurate result return.

The invention provides an intelligent analysis method, system, equipment and program product for automatic code review, which are applied to the field of computer software engineering and development tools, and the method comprises the following steps: obtaining code difference data, preprocessing to generate a semantic code unit, constructing a structured cue word and inputting the structured cue word into a large language model, according to the scheme, the problems that in existing code review, semantic understanding is insufficient, and efficiency is low can be solved, automatic risk identification and analysis of code changes are achieved, and review efficiency and accuracy are improved.

The invention discloses a method for automatically capturing semantic information of a binary code block and a related device, and relates to the technical field of computers.The method comprises the following steps that code disassembling and instruction level analysis are conducted on the binary code block, and a binary stream original instruction sequence is generated; performing semantic intermediate representation conversion and vocabulary overflow standardization on the binary stream original instruction sequence to obtain a binary code standardized instruction sequence; obtaining an instruction data dependency relationship and an instruction control dependency relationship corresponding to the binary code block, and constructing a corresponding binary code program dependency graph; semantic code node embedding is carried out on the binary code program dependency graph to generate a binary code representation vector corresponding to context semantics; and performing automatic semantic capture reasoning on the binary code representation vector corresponding to the context semantics through a pre-trained semantic understanding model to obtain semantic information of the binary code block. According to the method, the semantic information of the binary code block can be efficiently and accurately captured.

The invention discloses a multi-modal data quality evaluation method based on deep learning. The method comprises the steps of obtaining first modal record data and second modal record data which are marked to be aligned in a database; respectively carrying out semantic feature extraction on the first modal record data and the second modal record data to obtain a text structured data semantic coding feature tensor and a metadata description semantic coding feature vector; performing cross-modal joint coding on the text structured data semantic coding feature tensor and the metadata description semantic coding feature vector to obtain a structured-metadata cross-modal quality evaluation joint coding feature tensor; generating reconstruction record data based on the structured-metadata cross-modal quality evaluation joint coding feature tensor, and calculating offset features of the reconstruction record data; performing quality evaluation on the offset features through a first deep learning model; according to the method, through semantic coding feature extraction, fine-grained hash cross-modal coding and bidirectional reconstruction offset evaluation, the accuracy of data quality evaluation is improved.

The invention belongs to the technical field of computers, and provides a video virtual fitting method and device based on a diffusion model and a program product. The method comprises the following steps: firstly, acquiring a target clothing graph and a model video, and extracting depth features, masks and a model latent feature video; meanwhile, semantic codes of the clothes are extracted through a CLIP encoder. The first diffusion model processes the clothing graph and the CLIP code, and outputs a clothing space attention feature vector fusing fine-grained texture and semantics; and the second diffusion model serves as a video diffusion network, receives random noise, character features, masks and depth features, injects garment vectors and CLIP codes into the middle layer, and finally outputs a fitting video with naturally fitted and dynamically coherent garments. According to the method, the high-fidelity virtual fitting video can be efficiently generated according to the new clothing and model video, and the method has wide application value in the fields of electronic commerce and the like.

The invention relates to a rapid reasoning semantic communication method and system based on non-autoregressive decoding. The method comprises the following steps: step 1, constructing a semantic communication model of non-autoregressive decoding based on a neural network; 2, sequentially inputting the training batches into a semantic encoder and a source-channel joint encoder to obtain a semantic encoding matrix; step 3, obtaining a disturbed semantic coding matrix; 4, sequentially inputting the obtained disturbed semantic coding matrix into a source-channel joint decoder and a non-autoregressive semantic decoder to obtain a plurality of output probability matrixes; 5, training the semantic communication model of the non-autoregression decoding by using a cross entropy loss function, and finishing the training of the semantic communication model of the non-autoregression decoding when a preset number of times of training is reached; and step 6, reasoning through the trained non-autoregressive decoded semantic communication model, carrying out semantic communication, and obtaining a semantic reconstruction sentence. According to the method, the reasoning speed of the deployed semantic communication system is greatly improved, and the communication time delay is reduced.

The invention relates to the technical field of animal behavior analysis, and discloses an animal behavior accurate recording and analyzing system based on multi-mode keyboard coding. A data acquisition and synchronization module of the system acquires original key event data and generates a time sequence behavior data stream with millisecond-level precision; the coding logic engine module is used for converting the physical key sequence into a hierarchical behavior semantic coding sequence through a dynamic mode switching and key mapping technology; the behavior state modeling module is used for constructing an animal behavior state transition probability model, performing state decoding and estimation by adopting a hidden Markov model, and outputting a behavior state sequence with confidence; the behavior analysis and prediction module is used for predicting a behavior evolution trend and identifying an abnormal behavior event; and the threshold value dynamic adjustment module is used for dynamically adjusting an early warning threshold value through interaction between the normal behavior generator and the real-time discriminator aiming at an abnormal behavior event so as to finish accurate recording and risk analysis of animal behaviors.

The invention relates to the technical field of artificial intelligence, can be applied to the fields of financial science and technology and medical science and technology, and discloses an image counting method and device based on natural language guidance, equipment and a medium. Respectively extracting visual features and language features through an image coding module and a text semantic coding module; inputting the two features into a visual language unified embedding module for multi-modal fusion to obtain multi-modal features; performing semantic alignment screening on the multi-modal features through a prompt alignment visual filter to obtain target recognition features; inserting a counting mark into the target recognition feature, and inputting the target recognition feature into a quantity sensing decoding module for decoding to generate a natural language text containing a target quantity; and finally, outputting a natural language quantity answer text conforming to the grammar specification through a language generation header module. According to the method, the counting accuracy is remarkably improved through language-guided visual feature screening and a structured quantity decoding mechanism.

The invention discloses a multi-modal visual language navigation method based on a dynamic environment knowledge graph and related equipment, and can be applied to the technical field of artificial intelligence. After the real-time environment image of the area to which the target robot belongs is captured through the single camera, feature extraction is carried out on the real-time environment image to obtain the image features, the to-be-processed category corresponding to the target object is coded to obtain the semantic code, and after the spatial features of the target object are extracted, the target robot is obtained. Constructing a dynamic environment knowledge graph according to the confidence, semantic coding and spatial features of the target object in combination with an external knowledge base, extracting high-dimensional graph features of the dynamic environment knowledge graph, and performing implicit fusion and display modeling in combination with image features and language features corresponding to the target instruction to obtain a navigation state vector containing long-term memory; therefore, the moving operation of the target robot can be controlled based on the navigation state vector, efficient and explainable navigation reasoning is further realized, and the language navigation accuracy is improved at relatively low cost.

The invention relates to the field of body intelligence, and provides an autonomous decision-making method and system for a body intelligent robot. The autonomous decision-making method comprises the following steps of: acquiring environment and interaction data in real time by adopting an active perception mechanism, and converting the environment and interaction data into a scene and an own semantic code fused with the scene and the own; inputting the subject semantic code into a shunting decision-making layer to output a structured semantic abstract and an execution action, wherein the structured semantic abstract comprises scene parameters, user requirements and confidence; in response to the fact that the confidence is located in the middle confidence interval, determining and executing a non-invasive tentative action instruction used for verifying the user demand based on the execution action and the scene parameters; collecting user feedback data when the non-invasive heuristic action instruction is executed to determine a user feedback result; and determining a decision result based on a user feedback result, and adjusting the range of the dynamic confidence interval and the association weight between the execution action in the shunting decision layer and the user demand.

The application belongs to the technical field of computers, and provides a video virtual try-on method, device and program product based on a diffusion model. First, a target garment image and a model video are acquired, and a deep feature, a mask and a model latent feature video are extracted therefrom; meanwhile, a semantic code of the garment is extracted through a CLIP encoder. A first diffusion model processes the garment image and the CLIP code, and outputs a garment spatial attention feature vector that fuses fine-grained texture and semantics; a second diffusion model serves as a video diffusion network, receives random noise, a person feature, a mask and a deep feature, and injects a garment vector and the CLIP code into an intermediate layer, and finally outputs a try-on video that is natural and dynamically coherent. The application can efficiently generate a high-fidelity virtual try-on video according to a new garment and a model video, and has wide application value in the field of e-commerce.

The application discloses an attention mechanism-based adversarial text defense method and system, which comprises the following steps: inputting to-be-recognized text into a natural language processing model with an encoder-decoder as a basic structure; calculating the importance score of each word in the text by using a word scoring function; taking the reciprocal of the importance score to form a reconstruction score vector; calculating the weight of each hidden layer vector according to an attention formula to obtain an attention weight vector; balancing the reconstruction score vector and the attention weight vector by using the way of multiplying the reconstruction score vector by a hyperparameter, multiplying corresponding elements in the reconstruction score vector and the attention weight vector one by one to obtain a final reconstruction attention vector; and multiplying the reconstruction attention vector by a hidden layer feature vector to obtain a reconstructed semantic code, and obtaining output after decoding. The application has good generalization performance, does not need to retrain a model when coping with new adversarial attacks, and has a certain effect on character-level adversarial attacks and word-level adversarial attacks.

This invention discloses an autonomous data processing system based on a deep mining framework, relating to the field of data processing technology. It includes: a semantic judgment module that extracts the target missing field, associated fields, and historical homologous field sequences from input records; determines the field constraint relationships between the target missing field and associated fields; and constructs a field constraint graph accordingly; calculates the missing position value, constraint break value, and historical offset value; and determines the missing semantic code based on these values; and writes the field constraint graph, the missing semantic code, and the missing position value, constraint break value, and historical offset value into a record-level state payload. This invention constructs a continuous data processing link around the target missing field, associated fields, and historical homologous field sequences; and establishes clear data flow and control flow relationships between vectors such as field constraint relationships, field constraint graphs, missing semantic codes, original data tracks, and completed data tracks.

The present disclosure provides a semantic communication method and device, relates to the technical field of communication, in particular to the technical fields of large model, deep learning, diversity processing and the like. The specific implementation scheme is: obtaining source data; determining a semantic code symbol sequence and an importance ranking sequence number of a semantic code symbol in the semantic code symbol sequence based on the source data; obtaining a target code symbol sequence based on the importance ranking sequence number and the semantic code symbol sequence; and transmitting the target code symbol sequence to a receiving end through different independent fading paths, so that the receiving end obtains target data corresponding to the source data.

The invention discloses a bimodal combined ship electronic chart system and method based on a Beidou satellite, and belongs to the technical field of ship electronic charts. The method comprises the following steps: collecting a ship multivariate time sequence and key equipment image information, and integrating the information in a terminal server; singular term elimination is carried out on the ship multivariate time sequence, and the singular term is provided for the Beidou satellite information transmission terminal module; extracting a key structure and transmitting an image to a ship operation and maintenance image semantic coding module; a PearSGAN model is established; processing the bimodal information by using a Beidou coding structure; and executing an improved Mercator projection transformation method based on an elliptic dynamic equation, and providing a processing result to a centralized control center high-definition display screen module. The ship multivariate time sequence and the image information are deeply combined to form bimodal combined input information, and a high application value and a general paradigm are provided for further development of navigation guarantee, comprehensive management and other technologies of ocean ships in China.

The invention discloses a cross-platform virtual asset intelligent identification and energy efficiency management system based on the Internet of Things, and the system comprises an Internet of Things edge access and preprocessing module which is used for collecting operation parameter data; the structured Prompt construction module is used for generating a structured Prompt template; the SigLIP model double-tower semantic coding module is used for generating a virtual asset semantic vector set; the virtual asset identification and index mapping module is used for extracting an energy consumption time sequence; the improved TimesNet modeling module is used for executing periodic convolutional coding, energy consumption mode decoupling and cross-period dependence modeling, and outputting a cross-period context feature tensor; the prediction layer module is used for generating an energy consumption prediction result; and the energy efficiency strategy module is used for forming closed-loop control. The method has the advantages that the energy consumption trend is accurately predicted, the cross-cycle energy efficiency optimization strategy is realized, and the intelligent level and response efficiency of energy consumption management are greatly improved.

The application discloses a hydrological remote sensing image target recognition method based on a deep semantic model, relates to the technical field of remote sensing image processing, and comprises the following steps: obtaining a primary semantic code of a hydrological remote sensing image through a deep semantic network, calculating a semantic attribution probability distribution according to the primary semantic code, generating an initial attention guide signal, dynamically adjusting the weight of a specific perception path in the network, realizing first feature re-extraction to obtain refined features, calculating a feature compensation vector based on the confidence deviation generated by matching the features with a standard feature library, correcting the refined features to obtain enhanced features, and inputting the enhanced features into the network again for analysis to obtain a final recognition result. Through the attention adjustment of the semantic guide and the feature compensation based on the confidence feedback, the method effectively improves the recognition accuracy of complex hydrological ground object targets and the model robustness.

The invention relates to a radio intelligent understanding method and device based on dual-channel semantic coding deep learning, belongs to the technical field of dual-channel radio signals, and solves the problem of low signal understanding accuracy caused by lack of high-latitude associated information in the prior art. The method provided by the invention comprises the following steps: denoising a target dual-channel radio signal; performing high-dimensional semantic coding on the de-noised dual-channel radio signal to obtain a semantic coding signal of the dual-channel radio signal; and inputting the semantic coding signal of the dual-channel radio signal into a trained classification model for classification and identification, and obtaining the type of the dual-channel radio signal. According to the invention, the high-dimensional information is utilized to realize the understanding of the wireless signal, and the accuracy of signal understanding is improved.

The present disclosure provides a video recommendation method, device, electronic device and medium, which relate to the field of artificial intelligence technology, especially to the fields of big data, intelligent recommendation, large models and deep learning. The method includes: obtaining at least two candidate videos that match the target object; determining a first semantic code group corresponding to the candidate video; wherein the first semantic code group is used to describe at least two levels of semantics of the candidate video; obtaining a semantic interest tree pre-constructed for the target object; wherein the semantic interest tree is constructed according to input operations corresponding to historical videos and a second semantic code group corresponding to historical videos, and the historical videos are videos that the target object has viewed; based on the first semantic code group and the semantic interest tree, determining the recommendation order between the candidate videos; and performing the video recommendation operation according to the recommendation order. The present disclosure effectively improves the accuracy and personalization of video recommendations, and using the present disclosure for video recommendation is conducive to improving user experience.

Embodiments of the present application provide a text processing method, device and computer storage medium. The method comprises: obtaining a text to be processed; calling a first classification network to process a secondary category code representation, category matching parameters and semantic code representation of the text to be processed, to obtain classification information of a secondary category of the text to be processed; and determining a classification result of the text to be processed based on the classification information of the secondary category and classification information of a primary category of the text to be processed, the classification information of the primary category of the text to be processed being obtained by processing the text to be processed by a second classification network, so that the accuracy of hierarchical classification of the text can be improved.

This invention discloses a behavior anomaly detection method for bank personnel based on behavioral sequences, belonging to the field of behavior anomaly monitoring technology. It addresses the technical problems of existing methods for comprehensively monitoring and identifying high-risk behavioral characteristics of bank employees, which suffer from isolated behavior detection and a lack of semantic association. The method includes: parsing and aligning multi-source heterogeneous data collected from various data sources; performing common baseline analysis of group behavior under standard behavioral sequences on the behavioral path data of bank personnel to obtain standard role behavior baselines; performing semantic code mapping processing on behavioral metadata related to real-time behavior serialization to obtain semantically annotated real-time behavior code strings; performing dual evaluation calculations on the real-time behavior code strings based on path deviation and temporal deviation to obtain a comprehensive risk score; and automatically fixing the evidence chain of high-risk associated behavioral sequences with outliers in the comprehensive risk score to generate anomaly behavior detection results.

The invention discloses a semantic sentiment analysis method, an intelligent cabin, a medium and a vehicle. According to the scheme, user voice is split, and global semantic information is put into a token sequence for coding; a target lexical element semantic code of each independent neural network is screened from the target lexical element semantic codes by using a gating network, and is processed into an emotion semantic feature; newly adding a target neural network, and generating shared semantic features in combination with the gating network; therefore, shared semantic features, global semantic codes and own emotion semantic features are fused in each independent neural network, when a specific emotion category is analyzed, local emotion semantic features are focused to serve as a detail basis, and overall emotion deviation can be grasped by means of the global semantic codes, so that the emotion classification accuracy is improved. In addition, semantic support common with all emotion categories can be obtained by sharing semantic features, and deep mining is performed on local and overall association, so that judgment of each emotion category can give consideration to detail expression and global emotion deviation, and the emotion analysis ability of the intelligent cockpit is improved.

The application provides a multi-modal database query engine and method based on unified semantic representation, the engine is associated with a vector library of entity unified semantic vectors, and the core comprises five modules: a query analysis and semantic coding module processes single-modal and multi-modal queries, analyzes semantic structures, and generates corresponding semantic vectors or fused semantic vectors; a query execution plan generation module generates an optimal plan containing a search strategy, weight distribution, and resource scheduling according to semantic characteristics, vector library characteristics, and the like; a multi-modal query execution module obtains candidate results from the vector library according to the plan; a query result fusion module performs weighted fusion on the multiple candidate results, and generates a weighted score of each candidate result; and an output module outputs the results in order of the weighted scores. The engine realizes efficient processing and accurate result returning of multi-modal queries.

This invention discloses an AI-driven DevOps adaptive fault repair method and system. The method specifically includes: extracting semantic features from code and logs generated during system runtime to generate semantic code fingerprints; using the semantic code fingerprints as initial node features to construct an error propagation graph based on the software system's call relationships; utilizing graph convolutional networks to perform high-order topological feature learning on the error propagation graph to accurately deduplicate error information and locate the root cause of the fault; generating elastic scaling strategies and resource configuration optimization schemes for cloud-native clusters based on the deduplicated error information and the system's real-time load pattern; and generating corresponding repair plans and matching the optimal responsible party based on the located root cause of the fault and the real-time status characteristics of team members. This invention automates the entire DevOps fault repair process, significantly reducing the mean time to recovery and improving system stability and operational efficiency.

The invention relates to the technical field of BOM management in a PLM system, in particular to an EBOM and PBOM intelligent matching method based on Tree-LSTM and Transform fusion, which mainly comprises a data preprocessing module, a structure coding module, a semantic coding module, a double-model feature extraction module, a fusion layer training module and a matching output module. All the modules are connected in sequence to form a complete deep learning matching process. The method comprises the following steps: firstly, in a data input stage, a system exports EBOM and PBOM bills of materials from a PLM system; each table comprises typical fields, such as part numbers, part names, father node numbers, child node numbers, the number, units and other information. Each input sample is composed of a pair of nodes to be matched, namely (EBOM node and PBOM node). Each node contains text attribute fields (name, specification, description and the like) and structure attribute fields (hierarchy depth, father-son relation and the like).

The invention discloses an exhibition hall natural language positioning method and system based on a dual-path spatial semantic field and a storage medium. Performing context coding on the geometric layout of the booth through a layout coding network to form a learnable space base; a text encoder and a posture encoder are used for encoding natural language questions input by a user and audience postures respectively, the natural language questions and the audience postures are fused into unified semantic representation through a fusion network, and anchor semantic codes, relation semantic codes and direct semantic codes used for spatial reasoning are generated; then executing two reasoning paths of an anchor path and a direct path in parallel; and finally, fusing the outputs of the two paths in a probability domain to obtain booth probability distribution, and synchronously outputting a target booth ranking, anchor point interpretation information and a continuous thermodynamic diagram visualization result. According to the method, joint modeling is carried out on natural language questions, audience postures and geometric layout of booths in an exhibition hall scene, so that directional, reference and range question positioning is realized.

The invention discloses a proxy dynamic address and identity management system based on cache capability, and the system comprises a capability feature extraction module which is used for obtaining a capability attribute set of a proxy in real time, carrying out the denoising and normalization of the capability attribute set, and generating a standardized capability feature set; and a high-dimensional semantic coding module connected with the capability feature extraction module and used for mapping the standardized capability feature set to a high-dimensional feature space by using a preset semantic coding model to generate a high-dimensional feature vector. Perfect balance is achieved between data timeliness and resource efficiency through the system, the updating strategy periodically scans capacity change through a fixed time window, and the passive updating strategy can respond to model upgrading or hardware recovery emergency state events. The combination of the two ensures that the data in the edge node cache is always kept synchronous with the proxy body state of the physical world, and the storage space is prevented from being fully occupied by invalid historical data through a first-in first-out strategy.