23 results about "Semantic property" patented technology

A software package is received and unpacked into multiple components comprising plural functions. Each function is lifted from machine code into static single-assignment intermediate representation and tokenized to produce semantics-preserving embeddings. Intermediate-representation data-flow features are extracted, including detection of constant static variables on a stack, stack reaching definitions, uninitialized variables, and intra-procedural aliases. For each component, the embeddings and features are input to a machine-learning model trained on semantic properties derived from a corpus of software packages to generate a software supply chain risk level. Data characterizing the risk level is provided to a consuming application. When the risk level satisfies a remediation criterion, a remediation action is initiated, including generation of a source-code patch recommendation for an identified root-cause function, insertion of a runtime guard into the component, or issuance of a security advisory for distribution to a security operations dashboard.

The embodiment of the invention provides a code error repairing method based on a compiler, which comprises the following steps: receiving a target code written by using a target language, and executing a first repairing operation on the target code in the process of compiling the target code by the compiler to obtain a first repairing sequence, the first repairing operation comprises lexical error repairing corresponding to the lexical analysis process and grammatical error repairing corresponding to the grammatical analysis process. Performing a plurality of rounds of semantic repair operation based on an abstract syntax tree, wherein the abstract syntax tree used for the first round of semantic repair operation is generated after the first repair sequence is subjected to syntactic analysis by the compiler; any round of semantic repair operation comprises the steps of cloning a current abstract syntax tree to obtain a cloned syntax tree, and performing semantic attribute labeling based on the cloned syntax tree; and for any error node with the error attribute label in the clone syntax tree, performing semantic repair on a node corresponding to the abstract syntax tree.

The invention discloses a video key frame extraction method and device, computer equipment and a storage medium, belongs to the technical field of artificial intelligence, and is applied to key frame extraction of an insurance video or a healthy old-age care video clip. Structured decoupling of static and dynamic semantic features is realized by introducing a hierarchical variational auto-encoder, so that the model can be independently modeled and processed for different semantic attributes, and the semantic resolution capability of key frame extraction is remarkably improved. In combination with a semantic consistency measurement mechanism, the semantic deviation degree of the enhanced sample is evaluated and screened in real time in a potential space, it is ensured that the spatial relation and time logic of key visual elements are not damaged in the enhancement process, and semantic rationality and training stability of the enhanced sample are ensured. Besides, the performance feedback and the sample diversity state in the training process are dynamically perceived through the reinforcement learning strategy controller, the potential spatial disturbance intensity is automatically adjusted, and closed-loop self-adaptive optimization of the enhancement strategy is achieved.

The invention discloses an advertisement material intelligent optimization system and method based on knowledge extraction and knowledge fusion, and the method comprises the following steps: S1, obtaining and preprocessing a to-be-optimized advertisement material, and collecting an external knowledge source; s2, combined semantic analysis is executed, and cross-modal semantic alignment is carried out; s3, extracting entity units and corresponding semantic attributes, performing type labeling on the entity units, and establishing a semantic relation structure; s4, performing semantic fusion with an external knowledge source, calculating similarity between semantic vectors, performing entity mapping, performing normalization, and correcting conflict boundaries and attribute value divergence; s5, extracting a target expression fragment, and performing optimization according to a preset optimization rule set; and S6, verifying and detecting the optimized advertisement material, and outputting an optimization result. According to the method, deep understanding and cross-modal fusion optimization of advertisement materials can be realized, and the accuracy, attraction and propagation effect of advertisement semantic expression are improved.

A reasoning implementation method and apparatus based on a labeled property graph data model, and a computer device. Specifically, the reasoning implementation method based on a labeled property graph data model comprises: extracting a keyword included in a user inquiry and a semantic property associated with the keyword (S101); on the basis of the semantic property, converting the user inquiry into a second inquiry (S102); and on the basis of the second inquiry, providing a reasoning answer based on a labeled property graph data model (S103). The reasoning implementation method based on a labeled property graph data model can extend semantics of a labeled property graph model, so as to use the labeled property graph data model to provide more accurate and rational reasoning answers to user inquiries; and the disclosed method is relatively simple and fast to implement.

The invention discloses a zero sample fault detection method based on weighted semantic consistency embedding, and belongs to the field of industrial fault diagnosis. The method comprises the following steps that firstly, industrial equipment is used for collecting multi-mode sensor data, and standardized data are obtained through normalization, wavelet denoising and time sequence segmentation preprocessing; determining a fault attribute dimension based on domain expert knowledge, assigning each fault class to form a semantic attribute vector, and constructing a semantic attribute weight matrix through mutual information gain between attributes and fault tags; embedding data and semantic attributes into a shared space by using a data encoder and a semantic attribute encoder, decoding and calculating reconstruction loss by combining a data decoder and a semantic attribute decoder, and introducing weight to construct weighted semantic consistency loss; combining the two types of loss to train an encoder and a decoder, and extracting training data features by using the trained encoder and training a multi-attribute classifier; and finally, in a zero sample scene, extracting embedded features without fault data, inputting the embedded features into an attribute classifier, and matching fault categories through a maximum posterior probability. According to the method, the equal-weight hypothesis defect of an existing semantic consistency embedding method is broken through, key semantics are strengthened through mutual information gain weights, consistency loss is weighted to resist noise interference, and the precision, robustness and generalization ability of zero sample fault diagnosis are remarkably improved.

The invention discloses a multi-subject semantic attribute binding image generation method based on a diffusion model. The method comprises the steps of improving independent coding of a single sentence into segmentation of the sentence into a plurality of sub-sentences and independent coding of the sub-sentences; improving a denoising process and improving a cross attention mechanism; clear semantic representation is obtained by independently coding clauses, a denoising process is divided into a reconstruction branch and a generation branch, intermediate potential representation is spliced and fused to improve consistency and visual quality, and meanwhile, independent attention guidance is realized by combining an improved cross attention mechanism and mask division, so that feature aliasing is avoided, and the accuracy and the reliability of the system are improved. And the local precision and the space decoupling capability are enhanced. According to the multi-entity semantic modeling and image-text alignment method, a finer and more stable solution is provided while high efficiency and universality are ensured, and the potential in multi-entity semantic modeling and image-text alignment is shown.

The invention relates to the technical field of image rendering, and discloses a semantic scene rendering method based on three-dimensional Gaussian representation. The method is used for solving the problems that in a traditional method, artifacts are likely to be generated under high-frequency texture and strong perspective, semantic output is not stable, and noise is large. The method comprises the following steps: firstly, acquiring a multi-view frame, registering scale gears and shielding candidate bands, constructing a Gaussian set, registering mass components such as geometry, appearance, semantic attributes, view angle consistency, scale consistency, boundary consistency, time consistency and the like, and generating a boundary protection range and a visibility projection cluster; the method comprises the following steps: selecting candidates for a target view angle according to a quality threshold, resolving projection competition, generating an adaptive set, simultaneously being compatible with appearance and semantics in the same session, sequentially performing scale suppression, perspective consistency and occlusion gating on a semantic side, performing write-back on related quality components and version identifiers by combining cross-view angle verification and noise budget, and keeping semantic output consistent.

The application provides a privacy record linkage method and system for semantic heterogeneous data, involving multiple linkage participants and a linkage unit. Firstly, the attribute is divided into strong and weak categories by calculating the average difference of semantic similarity of data attribute characters. For weak semantic attributes, an improved CLK-CRBF counter Bloom filter is used to generate integer encoding. For strong semantic attributes, a fine-tuning embedding model is used to generate a dense vector. Then, the integer encoding is fused as a structured disturbance with the dense vector to form a semantic-aware encoding vector and sent to the linkage unit. The linkage unit uses locally sensitive hashing in an adaptive hybrid encoding space to block, generate candidate record pairs in the group, and calculate the similarity. Finally, the matching result is determined according to the negotiated threshold. The method significantly improves the linkage effect of semantic heterogeneous data while ensuring privacy, and the greater the data difference ratio, the more obvious the advantage.

According to the multi-object video editing method based on semantic perception content alignment provided by the invention, a precise semantic-level editing effect with consistent time sequence can be realized in a multi-object video without training a diffusion model. Effective semantic alignment is realized through two core mechanisms of semantic adaptive modulation and semantic prior modeling. Specifically, coupling semantic features among multiple objects are effectively decoupled through an attention regulation and control mechanism of region perception and text perception, so that the model can accurately position the spatial position and semantic attribute of each object; meanwhile, cross-modal semantic consistency supervision is constructed based on a pre-trained visual language model, and the diffusion sampling process is guided to keep global semantic consistency with text editing prompt. Experimental results show that the method is remarkably superior to the prior art in the aspects of semantic alignment, editing precision and visual consistency, can be widely applied to the fields of film and television production, human-computer interaction, digital media creation and the like, and has excellent practical value.

The invention discloses a position privacy protection method and device based on semantic indistinguishability, equipment and a medium, and relates to the technical field of privacy protection. The method comprises the steps of firstly obtaining an interest point library, a semantic attribute library and historical movement data of a user; according to the interest point library and the semantic attribute library, obtaining a plurality of interest points similar to the semantic attributes of the actual positions as observation positions; forming an anonymous set by the actual position and the plurality of observation positions; according to historical mobile data, judging whether the anonymous set meets semantic indistinguishable constraints or not by adopting a forward algorithm of an HMM (Hidden Markov Model); a server is accessed using an anonymous set that meets semantic indistinguishable constraints. According to the method and the device, the inference attack based on the behavior pattern is simulated, so that an attacker cannot distinguish the impending (or current) behavior of the user by the anonymous position track with semantics indistinguishable, and the privacy risk caused by observing the release track is eliminated, so as to achieve the effect of resisting the attack.

The invention provides a tool breakage state monitoring method and device based on zero sample learning and a medium, and the method comprises the steps: generating a high-dimensional semantic vector according to a word vector corresponding to a tool-breakage structured text description and a graph embedding vector corresponding to a tool-breakage knowledge graph; generating a multi-modal sensing feature vector according to the data of the to-be-detected cutter in the cutting process; mapping the multi-modal sensing feature vector to a semantic space corresponding to the high-dimensional semantic vector; generating a virtual sample by taking the high-dimensional semantic vector corresponding to the sample of the invisible damaged category as a condition; training a zero sample classifier according to the virtual sample; the dynamic threshold value is determined according to the zero sample classifier, real-time early warning monitoring is conducted on the to-be-detected tool based on the dynamic threshold value, high-precision recognition of unknown damage categories can be achieved by means of priori knowledge, namely semantic attributes, of the tool and data of known damage modes, and therefore the adaptive capacity of a monitoring system in the variable machining environment is improved.

The application discloses a lung nodule detection and semantic attribute rating method based on multi-task learning, which comprises two sub-tasks of lung nodule detection and semantic attribute rating. In order to realize feature sharing of joint learning between the two sub-tasks, the application connects a lung nodule detection sub-network and a semantic attribute rating sub-network together to form an end-to-end joint model. The lung nodule detection sub-network acquires position information of the nodule, and the output thereof serves as input of the semantic attribute rating sub-network. The two sub-networks share bottom layer features in a down-sampling stage of a U-Net network, so that feature sharing of the multi-task model is realized. In the process of joint learning training of the two tasks, since the training difficulty and convergence speed of different sub-tasks can be different, the application adopts a dynamic weight average method to adjust loss weights of different tasks. The method can not only effectively detect lung nodules, but also identify semantic attributes of the lung nodules as an additional supervision signal.

This invention discloses a semantic attribute set generation method for zero-shot learning. Through multi-stage collaborative training of a teacher model and a student model, the accuracy and consistency of the generated attribute set are improved. The teacher model, based on a conditional generative adversarial network, generates a high-quality image I1 using an attribute set constructed by an expert. The student model, based on an attention mechanism and a visual Transformer model, extracts key features from the original image to generate a preliminary semantic attribute set and a corresponding image I2. Using the image I1 generated by the teacher model as a benchmark, I2 is compared with I1 to obtain the feature differences between them. These feature differences are used as guiding signals to back-optimize the semantic attribute set A2 while simultaneously optimizing I2. Finally, the student model generates a semantic attribute set A that is consistent with the expert-constructed attribute set A1. f The innovation of this invention lies in its ability to generate high-quality semantic attribute sets in scenarios where expert attribute sets are unavailable. This addresses the problem of zero-shot learning being unsuitable due to gaps in expert attribute sets for certain categories, thus expanding the application scope of zero-shot learning methods.

The application provides a small sample fabric defect detection method based on decoupling variational prototype, comprising: constructing a small sample fabric defect detection network model, including a backbone feature extraction network and a variational prototype learning network based on semantic decoupling and orthogonal decoupling which are connected in sequence; introducing a semantic decoupling variational prototype module SDVP and a semantic anchoring orthogonal decoupling constraint module SODC in the variational prototype learning network VFA, the semantic decoupling variational prototype module SDVP is used for multi-attribute semantic decoupling and constructing a global semantic alignment loss to eliminate the mutual coupling of multi-attribute visual features, and the semantic anchoring orthogonal decoupling constraint module SODC is used for adopting a semantic anchoring and orthogonal constraint method to enhance the independence and discriminability between different semantic attributes, and using a small sample target detection task data set to train the fabric defect detection network model and apply it to defect detection; and the application can fully mine the fine-grained semantic information of the fabric defect under the condition of limited samples, and enhance the discriminability and independence between different semantic attributes.

The invention relates to the field of image processing, and discloses an image coding method and system, and the method comprises the steps: designing a semantic attribute structured hidden space; at a sending end, encoding original image data into corresponding text semantic description, decoupling parameter semantic description and a semantic attribute value set, and selecting required transmission data for sending according to feedback of channel parameters; after receiving the transmission data, a receiving end obtains corresponding hidden space data according to the semantic description information of different forms; and after the required hidden spaces with different semantic forms are obtained, fusing the hidden spaces with different semantic forms into a complete hidden space for image information reconstruction, and performing image information reconstruction according to the complete hidden space. The method has the advantages that image semantic high-fidelity reconstruction can be realized under an extremely low bit rate transmission condition; the robustness of a communication system can be effectively improved by using a cross-modal alignment mechanism; and adaptive switching from parameter semantic dominance to text semantic dominance can be realized.

The invention relates to the technical field of entity extraction, in particular to a general entity extraction method, a medium and equipment, and aims to convert isolated entity names into structured refined names of attributes and names and provide richer semantic information for entities by adding corresponding semantic attributes to names of initial entities. The context corresponding to the initial entity which does not pass semantic uniqueness verification is expanded and iteratively updated to supplement more semantic details, so that the newly generated refined entity name has higher semantic distinction degree, the semantic conflict problem caused by insufficient information is solved, and the semantic uniqueness of the entity is improved. It is ensured that the finally output target entity name has strict semantic uniqueness, and inherent defects of an initial entity are thoroughly overcome; and the corresponding target entity extraction model is constructed and trained for each entity name set, so that the model only focuses on a certain type of entities with specific semantics, and the extraction effect for the entities with specific semantic attributes is greatly improved.

The invention discloses a poultry industry supply chain-oriented time sequence demand perception method and system, and relates to the technical field of time sequence data analysis, and the method comprises the steps: carrying out the extraction from historical time sequence sales volume data and semantic attribute data through multi-modal coding and comparative learning, and carrying out the clustering to form a common pattern library with semantic tags; in a sensing stage, time sequence and semantic data of an object to be sensed are mapped to a unified space, and a reference common pattern is retrieved based on two stages of semantics and sequence similarity. Then, the personalized time sequence features are dominated, the generality mode is used as a reference, self-adaptive fusion is achieved through an attention mechanism, and feature representation is enhanced; and finally, generating a time sequence demand sensing sequence of a future target time period by using a decoder. The method has higher perception stability and interpretability in a data sparse and cold start scene, and can effectively support production scheduling, inventory and sales decisions in a poultry industry supply chain.

The invention discloses a synthetic zero sample learning method inspired by human cognition, which belongs to the field of zero sample learning and comprises the following steps: firstly, clustering low-level semantic attributes to automatically mine high-level semantic concepts; secondly, sensing and extracting fine-grained visual feature fragments related to the advanced concepts; and finally, adaptively fusing the feature fragments from different visible classes to synthesize a high-quality unvisible class sample. The whole process and classifier training are collaboratively optimized in a unified framework. According to the method, the recognition accuracy and generalization ability of the model to the unseen categories are remarkably improved.

The invention discloses a distributed data security processing method oriented to multi-source heterogeneous data, and relates to the technical field of data security processing, and the method comprises the following steps: S1, collecting multi-source heterogeneous data, analyzing grammar rules and semantic attributes of various types of data, determining format types, core semantic information and sensitivity levels of the data, and forming a data feature file; according to the method, comprehensive analysis of grammar rules and semantic attributes is carried out on multi-source heterogeneous data, an adaptive encryption scheme is automatically matched in combination with data features, efficiency loss or security vulnerability caused by a single encryption mode is avoided, the dynamic permission atlas is constructed based on access subject identity attributes, task scenes and historical behaviors, and the security of the access subject is improved. Real-time binding of the data desensitization level and the access permission is realized, the problems of encryption strategy solidification and mismatching of permission control and data heterogeneous characteristics in a distributed system are solved, and the data processing efficiency and the access flexibility are improved on the premise of guaranteeing the distributed processing security and integrity of the multi-source heterogeneous data.

The application discloses a large model text generation method based on semantic adaptation and dynamic hierarchical comparison, and belongs to the technical field of artificial intelligence and natural language processing, which can at least partially solve the problems of lack of semantic perception ability, fixed hierarchical selection strategy and waste of computing resources in the prior art contrast decoding technology, and introduces a semantic routing module in the decoding process.The method obtains the hidden state of different levels of the model and the final layer output distribution when the model reasons and generates each word element; the semantic routing module is used for semantic attribute determination and uncertainty calculation on the current generation state; according to the determination result, a strategy is dynamically selected from a preset decoding strategy set to calculate the final output value; and the final output value is normalized and combined with a sampling algorithm to generate a word element.The application realizes dynamic balance of factuality and fluency, suppresses hallucinations and reduces reasoning delay.

The application discloses an electronic component zero-sample recognition model and method based on visual information and semantic attributes, and the recognition model comprises a visual classification network model, a Skip-gram model and a CBOW model. According to the visual features of the electronic components, a semantic attribute table of visible class and invisible class electronic components is designed, so that the semantic attributes between the visible class and the invisible class are shared. By combining the visual information and the semantic attributes of the labels, the semantic attributes of the labels learned from the text data are used as the supervision signals, and the visual classification network is trained in combination with the seen class images, and then zero-sample learning is realized. During the training, the visible class is converted into the semantic attribute class through the Skip-gram model, so that the training process of the visual classification network becomes the training between the shared semantic attribute class and the input image. In the prediction process, the visual classification network predicts the shared semantic attribute class, the CBOW model converts the semantic attribute class into the class label, the prediction of the visible class and the invisible class is realized, and the zero-sample recognition of the electronic components is completed.

The application discloses a kind of based on knowledge graph and big language model's examination point accurate diagnosis method and system, belong to intelligent education technical field, this method includes: the subject knowledge graph comprising knowledge point and inheritance, composition, precursor relationship is constructed;Symbolic analysis and textbook progress constraint are carried out to input question, and preprocessing information is obtained;Parallelly execute full-text search, semantic attribute matching, fuzzy matching and, recall candidate knowledge point;Knowledge structured extension is carried out by parent class backtracking, subclass expansion;The application is coupled by the structured constraint of knowledge graph and the semantic understanding depth of big language model, effectively solve the term of general big model in education scene is not standard, implicit examination point is difficult to dig and illusion problem, accurate, interpretable examination point diagnosis is realized, and reliable technical support is provided for intelligent tutoring and individualized learning.