28 results about "Contextual similarity" patented technology

The present disclosure enhances domain lookalike detection by integrating a phishing risk assessment score into a multi-layered evaluation framework. The method systematically generates lookalike domains through genetic algorithms. Registered domains undergo advanced phishing analysis, incorporating domain and URL checks, technical infrastructure assessments, content inspections, and reputation-based intelligence to calculate a dynamic phishing score. A comprehensive risk score is then determined by merging phishing likelihood with business attributes, graphical / contextual similarity metrics, and domain registration patterns. Domains are categorized into predefined risk levels including phishing, registered, preventative, company-owned, or watchlist, with specific action recommendations provided for each category. The system generates prioritized alerts for high-risk domains, offering customers actionable intelligence to mitigate threats. By combining phishing-specific indicators with contextual evaluations, this solution improves detection accuracy, reduces false positives, and enables organizations to respond effectively to domain-based threats in real time, addressing evolving cybersecurity challenges.

Artificial intelligence models are trained with training datasets of known input / output values. Test datasets are used to evaluate the trained artificial intelligence models. Context mismatch between the training dataset and the test dataset can slow down the development of artificial intelligence models. The described systems and methods can identify context similar datasets for the purpose of training and testing an artificial intelligence model. In one embodiment, a context similarity detector can ingest and combine a training dataset and a test dataset and generate a context similarity score for the two. If the score is above a threshold, the datasets are similar, and the relevant artificial intelligence model can be trained with one and tested with the other.

The invention relates to the technical field of internet of things monitoring and data unification of logistics assets, in particular to a multi-system data unification logistics asset internet of things monitoring and processing system, which comprises an atlas construction unit used for collecting multi-source heterogeneous data streams in real time and aggregating to generate dynamic time sequence diagram snapshots; the heuristic quantization unit is used for taking the score as a pseudo tag; the time sequence diagram neural network unit is used for generating high-dimensional state vectors for the independent nodes; the model training unit is used for enabling the similarity of the high-dimensional state vector in a vector space to reproduce heuristic context similarity; the unified asset aggregation unit is used for resolving the high-dimensional state vectors and aggregating the high-dimensional state vectors into a unified asset view; the hidden event inference unit is used for generating a hidden event alarm and judging the state as a normal state when the abnormal inference score does not exceed an alarm threshold value; the technical problem that the business system identifier and the Internet of Things identifier can be aligned only through manual labeling is solved, and automatic association of heterogeneous data and construction of a unified asset view are achieved.

The present application relates to a kind of incomplete speech rewriting model construction method and device, method includes: based on span dependency and insertion dependency dependency modeling and using node link resolution mode obtains the dependency graph of incomplete speech rewriting text editing operation;Using GPT model, the context similarity feature and / or rewriting consistency characteristic of current incomplete speech sentence are calculated, the context similarity feature and / or rewriting consistency characteristic are used to enhance the interactive inference of incomplete speech rewriting;The dependency graph score feature is fused with the context similarity feature and / or rewriting consistency characteristic, and the final feature after the feature fusion is pushed to rewrite result based on. More abundant semantic features can be provided for parsing model, and the speech rewriting effect is improved.

The invention relates to the technical field of software analysis, and provides a code similarity detection method and device. The method comprises the following steps: determining context information of a query function according to a binary dependency graph of the query function; determining context information of each candidate function according to the binary dependency graph of each candidate function; determining context similarity between the query function and each candidate function according to the context information of the query function and the context information of each candidate function; according to the content of the query function and the content of each candidate function, determining the content similarity between the query function and each candidate function; selecting an initial similar function from the candidate functions according to the context similarity and the content similarity; and inputting the initial similarity function and the query function into the large language model to obtain a code similarity detection result output by the large language model. According to the code similarity detection method and device provided by the invention, the code similarity detection efficiency and accuracy are improved.

The invention provides a Text2SQL semantic caching method based on context and pattern matching, and relates to the field of natural language processing and database interaction.The method comprises the steps that user query is input into a semantic compression module, semantic significant keywords are extracted to obtain a candidate set, a vocabulary set is extracted and constructed through a large language model, and a database is obtained; obtaining a compressed query through a syntactic organizer; inputting the historical dialogue into a context encoder, and obtaining global context representation through encoding and two-stage attention; performing coarse-grained filtering on the compressed query to form a newest candidate set; performing fine-grained context matching on the global context representation and the newest candidate set to obtain context similarity; and comparing the context similarity with a preset threshold value, judging whether the cache is hit or not, if the cache is hit, interacting the hit SQL statement with a database, and returning a query result. According to the method and the device, the problems of low SQL multiplexing accuracy and high response delay are solved.

The application relates to the technical field of image processing, and provides an image processing method, a computer program product, a storage medium and an electronic device. The image processing method comprises the following steps: obtaining the self feature of a target in a to-be-processed image and the self feature of the target in a reference image; fusing the self feature of the to-be-processed target with the self feature of at least one other target in the to-be-processed image and / or the reference image to obtain the context feature of the to-be-processed target, and fusing the self feature of the reference target with the self feature of at least one other target in the to-be-processed image and / or the reference image to obtain the context feature of the reference target; calculating the context similarity of the to-be-processed target and the reference target according to the context feature of the to-be-processed target and the context feature of the reference target; and determining the final similarity of the to-be-processed target and the reference target according to the context similarity. Since the context information of the target is fully considered when the similarity is calculated, the finally calculated similarity has high precision.

The application discloses a mental health support dialogue generation method based on multi-modal input, including: 1. Obtain text data and preprocess; 2. Calculate the context similarity to obtain a high-quality example set; 3. Use a visual language model to extract key visual clues from images, including facial expressions, actions and emotions; 4. Integrate an external common sense knowledge base to understand the user's emotions and intentions; 5. Generate counseling strategies by using a large language model; 6. Context-based dialogue generation. The application enhances the performance of large language models in mental health dialogue by comprehensively utilizing visual information, dialogue context and counseling strategies, enabling the generation of more detailed and empathetic supportive replies, thereby significantly improving emotional recognition and response quality, and more comprehensively understanding and responding to the emotional state of users to effectively improve the effect of mental health support.

Systems and techniques are provided for streamlining a proposal for a remote assistance system of an autonomous vehicle (AV) based on contextual similarities in remote assistance sessions that are associated with a set of AVs. An example process includes receiving a request for remote assistance from an AV in a scene and identifying a context associated with the AV. The context includes at least one of contextual factor(s) of the scene, AV characteristics, and task(s) associated with the request. The example process includes comparing the context associated with the AV with one or more contexts associated with one or more sessions associated with a set of AVs, determining a similarity between the context associated with the AV and the one or more contexts associated with the one or more sessions, and determining one or more proposals for the request based on the similarity.

A real-time fraud detection system is disclosed that models financial transactions as evolving memory graphs using biologically inspired dendritic architectures. The system employs graph neural networks with edge decay and reinforcement mechanisms to retain behavioral context over time, enabling dynamic anomaly detection. An attention-based recall module reactivates latent fraud patterns based on contextual similarity. Optionally, quantum walk engines and entangled memory states are integrated to accelerate pattern discovery in high-dimensional data. The invention supports few-shot meta-learning, real-time inference, and deployment on neuromorphic or quantum-classical platforms, offering adaptive, scalable, and interpretable fraud detection across dynamic financial ecosystems.

The present invention makes it possible to calculate an index with high accuracy, even for an area with limited data. This quality estimation system includes: a context data generation unit that generates a plurality of pieces of context data from measurement data; a first index calculation unit that generates first index data for each context and for each area, using the context data; an identification unit that identifies a second context similar to a first context; a second index calculation unit that defines an area for which the first index data is not generated, among a plurality of areas, as an unmeasured area, and generates second index data for the first context by using the context data of the first context for an area other than the unmeasured area, and using the context data of the second context for the unmeasured area; and an output unit that outputs the second index data.

The embodiment of the invention provides a geographic entity information processing method and device, a storage medium and a program product. In the embodiment of the invention, for at least two pieces of geographic entity description information, a geographic semantic feature encoder, a distance feature encoder and a geographic context feature encoder are respectively utilized to encode names and address semantics, spatial positions and surrounding geographic environments; obtaining a geographic semantic similarity, a spatial distance feature and a geographic context similarity; and inputting the geographic semantic similarity, the spatial distance feature and the geographic context similarity into a classification head network, and outputting a classification result of whether the at least two pieces of geographic entity description information point to the same geographic entity or not. Therefore, by introducing the model to learn the incidence relation between different features, full utilization of multi-dimensional features such as geographic semantics, spatial positions and surrounding geographic environments is realized, feature fusion and accurate judgment are realized in the classification head network, and the accuracy and robustness of geographic entity matching are improved.

Systems and methods are provided for determining calls and contacts associated with calls that are contextually similar based on content of the calls. Data associated with a call include one or more utterances made by speakers and a set of values indicating relevance between content of the call and topic categories of the call. The disclosed technology generates a topic vector associated with a call and / or respective speakers of the call. The topic vector includes a multi-dimensional vector where each dimension corresponds to a topic category. The disclosed technology determines calls that are contextually similar by comparing angular distances between topic vectors. A search query receiver receives a search query that queries contacts and calls that are contextually similar to a given call and / or a speaker. The disclosed technology identifies calls with topic vectors that are within a predetermined angular distance.

The application provides a non-structured content similarity determination method and device and electronic equipment, which normalizes extracted content obtained from non-structured content to obtain normalized content; sets a label for the normalized content based on a specified feature extracted from the normalized content, performs semantic similarity matching on the normalized content with the set label to obtain a first matching result; performs relationship calculation of metric fusion on the normalized content with the set label based on context correlation of the normalized content to obtain relationship metric information; performs context similarity matching according to the relationship metric information to obtain a second matching result; and performs content similarity matching on the non-structured content according to the normalized content, the label set for the normalized content and the relationship metric information to obtain a third matching result. This method comprehensively utilizes multi-dimensional methods of content analysis to perform similarity matching on non-structured content, so that each dimension method can complement each other, and the accuracy of analysis is improved.

This application provides a method for protecting prompt words in large models based on differential privacy. By constructing a perturbation mapping table, differential privacy perturbation is applied only to the first token encountered in a multi-turn dialogue, and the mapping relationship is stored. Subsequent repeated tokens directly reuse the historical perturbation results, thus avoiding the accumulation of privacy budget with the increase of dialogue turns. Simultaneously, an attention-based context-aware utility function is combined to integrate the static embedding similarity of tokens with dynamic context similarity to maintain semantic consistency across turns. Finally, a two-stage bucketing index mechanism is adopted, first bucketing candidate tokens according to utility, and then fine-sampling within the selected buckets, effectively mitigating the long-tail effect under large vocabulary and improving the stability of perturbation selection. This application achieves a balance between privacy protection strength and semantic usability in multi-turn dialogues, and only requires perturbing the input prompt words without accessing the model's internal architecture, making it suitable for black-box large language model inference scenarios.

The invention discloses a causal class activation mapping-based region perception open set identification method, which comprises the following steps of: performing closed set training through a convolutional neural network by taking a training image of a known class as input, and generating class activation mapping reflecting causal contribution of a local region to a classification result based on region ablation analysis; the method comprises the following steps: in a training stage, extracting a region perception feature vector of a training image by utilizing a saliency map, and constructing a region prototype corresponding to a known category; in a test stage, region perception feature vectors are extracted from a test image, a region consistency score is calculated with the region prototype constructed in the training stage, and fusion judgment is carried out in combination with a global uncertainty score based on an energy model, so that known category identification and unknown category rejection are realized. According to the method, causal class activation mapping is introduced into an open set identification judgment process, so that the capability of distinguishing context similar unknown classes can be enhanced, and a regional interpretable basis is provided.

The present subject matter discloses techniques to identify linkable items and accordingly generating recommendations indicating such linkages. One or more work data objects, each linked with a corresponding work item, may be obtained. Further, one or more support data objects, each linked with a corresponding support item, may be obtained. The one or more work data objects and the one or more support data items may be encoded into work vectors and support vectors, respectively. A similarity score may then be estimated based on a quantitative comparative assessment of the work and support vectors. The similarity score may indicate a contextual similarity between each of the one or more work data objects and the one or more support data objects. Based on the similarity score, the one or more support items may be linked with the one or more work items.

The invention discloses a shield tunnel construction project knowledge recommendation method based on a situation hypergraph. The method comprises the following steps: S1, collecting data information of a whole implementation period of a shield tunnel construction project, and constructing a shield tunnel construction project knowledge graph ontology library according to the data information; s2, based on the shield tunnel construction project knowledge graph ontology library in the S1, performing shield tunnel construction project knowledge situation modeling according to shield tunnel construction project knowledge situation dimensions, so as to construct a shield tunnel construction project multi-dimensional situation knowledge hypergraph instance; s3, based on the multi-dimensional situation knowledge hypergraph of the shield tunnel construction project in the S2, embedding entity attributes and triples of the multi-dimensional situation knowledge hypergraph of the shield tunnel construction project in sequence; and S4, constructing a neighbor information aggregation model based on a graph attention mechanism, forming a shield tunnel construction project knowledge recommendation model, operating the shield tunnel construction project knowledge recommendation model, and finally realizing Top-K knowledge recommendation through matching of situation similarity.

An apparatus, method, system and computer-readable medium are provided for generating one or more descriptors that may potentially be associated with content, such as video or a segment of video. In some embodiments, a teaser for the content may be identified based on contextual similarity between words and / or phrases in the segment and one or more other segments, such as a previous segment. Text / characters may serve as a candidate descriptor(s). In some embodiments, one or more strings of characters or words may be compared with (pre-assigned) tags associated with the content, and if it is determined that the one or more strings or words match the tags within a threshold, the one or more strings or words may serve as a candidate descriptor(s). One or more candidate descriptor identification techniques may be combined.

Disclosed is an artificial intelligence-based matchmaking system (100) for suggesting compatible user profiles on a graphical user interface (GUI). The system (100) includes a user device (102) comprising an input unit adapted to receive one or more input data from a user and an output unit communicatively coupled with a server (104) through a communication network (106). The server (104) includes a processing unit configured to receive and store profile inputs comprising textual, categorical, or behavioral data such as occupation, interests, and intent parameters, process the profile inputs to generate a prioritized list of candidate profiles based on semantic correlation, contextual similarity, and inferred intent, display the candidate profiles sequentially on the GUI, receive gesture inputs indicating interest or skip actions and update a dynamic parameter-weighted preference model to adaptively reorder subsequent profiles in real-time. The present disclosure also relates to a method (200) for artificial intelligence-based matchmaking system (100).

The invention relates to a robot control method based on a multi-modal large model, and relates to the technical field of robots. The method comprises the following steps: acquiring multi-modal input data and extracting context features; analyzing the semantic features, and extracting an instruction core intention; performing constraint condition extraction on the user instruction text to obtain a key constraint condition; recursively decomposing the complex instruction into an atomic task sequence according to the instruction core intention and the key constraint condition so as to generate a time-space continuous action planning candidate set; outputting a cross-modal correlation feature in the joint embedding space; calculating a context similarity between the cross-modal association feature and the historical interaction data feature; screening out an optimal action planning candidate with the highest similarity from the action planning candidate set based on the context similarity; and controlling the robot to move based on the optimal action planning candidate. According to the invention, the decision-making ability of the robot in a complex scene can be improved.

Described systems and methods provide a context similarity detector configured to receive two or more datasets, combine the datasets into a combined dataset, and perform clustering on the combined dataset. Based on the clustering, the context similarity detector generates a context similarity score indicating a similarity between the datasets and compares the score to a threshold. Datasets having a context similarity score above the threshold may be identified as context-similar and may be used to improve training and evaluation of artificial intelligence models.

The application discloses a kind of knowledge graph pipeline multi-strategy entity disambiguation incremental alignment method and system.The method normalizes candidate entity and generates fingerprint, to block key in bucket generation candidate entity pair, comprehensive similarity is obtained by fusing character, semantic embedding and structural context similarity, when comprehensive similarity is not lower than class correlation threshold, similar graph is constructed and representative entity is determined, only incremental alignment is executed to change fingerprint set and historical fingerprint index, and output incremental change set and conflict details.Compared with full amount pairwise comparison, the application reduces the scale of candidate pair and improves the processing efficiency in continuous updating scenario.

The application provides a Text2SQL semantic caching method based on context and pattern matching, relates to the field of natural language processing and database interaction, and comprises the following steps: inputting a user query into a semantic compression module, extracting semantic significant keywords to obtain a candidate set, extracting and constructing a vocabulary set through a large language model, and obtaining a compressed query through a syntax organizer; inputting historical dialogues into a context encoder, obtaining a global context representation through coding and two-stage attention; performing coarse-grained filtering on the compressed query to form a latest candidate set; performing fine-grained context matching on the global context representation and the latest candidate set to obtain a context similarity; comparing the context similarity with a preset threshold to determine whether the cache hits, and if the cache hits, interacting with a database with the hit SQL statement and returning a query result. The application solves the problems of low SQL reuse accuracy and high response delay.

This invention relates to the field of deep learning technology, particularly to a Transformer optimization method and system based on a token graph-based query propagation mechanism. The method comprises an input processing layer, a dynamic graph construction layer, a propagation control layer, a computational optimization layer, a feedback adjustment layer, and an output integration layer. This invention captures the contextual similarity or relationship between tokens by constructing a token-graph (dynamic or static structure) and introduces a stochastic propagation of Q vectors mechanism. Only the seed token generates a complete Q vector, while the remaining tokens generate approximate Q vectors through graph propagation. Furthermore, K and V are projected only onto the central node, and a local error control feedback module is included. This solves the problems of computational resource constraints, memory limitations, and accuracy stability in long sequence processing of Transformers. It provides a practical and feasible technical path for its application in complex tasks such as long document understanding, multi-turn dialogue, and code generation.

The present application relates to the technical field of logistics asset internet of things monitoring and data unification, in particular to a multi-system data unified logistics asset internet of things monitoring processing system, comprising a graph construction unit, used for real-time acquisition of multi-source heterogeneous data streams, and for aggregated generation of dynamic time series graph snapshots; a heuristic quantification unit, used for taking scores as pseudo labels; a time series graph neural network unit, used for generation of high-dimensional state vectors for independent nodes; a model training unit, used for making the similarity of high-dimensional state vectors in vector space reproduce heuristic contextual similarity; a unified asset aggregation unit, used for solving high-dimensional state vectors and aggregating them into a unified asset view; and a hidden event inference unit, used for generation of hidden event alarms, and for determination of normal state when an abnormal inference score does not exceed an alarm threshold. The present application solves the technical problem that business system identification and internet of things identification need to be manually annotated to be aligned, and realizes automatic association of heterogeneous data and construction of a unified asset view.

The application discloses an event element automatic extraction system and method in the technical field of artificial intelligence natural language processing, comprising a model iteration module and a data iteration module; the model iteration module is used for iteratively training an event element extraction model based on a double-state Prompt according to a labeled data set, and is also used for obtaining new labeled data generated by a data iteration model and updating the labeled data set; the data iteration module is used for obtaining similar contexts similar to event element contexts in the labeled data set from unlabeled text data through similarity measurement, extracting event elements from the similar contexts through the event element extraction model based on the double-state Prompt trained by the model iteration module, and is also used for generating new labeled data according to the event elements and the corresponding similar contexts. The application can establish the connection between different event types under the condition that only a small amount of labeled data samples are available, and quickly and accurately extract event elements.