663 results about "Classifier (UML)" patented technology

The invention relates to the technical field of block chains, in particular to a smart contract vulnerability detection method and device based on multi-modal features, and the method mainly comprises the steps: training a meta-learning model in a dynamic adaptation module, and adjusting the global parameters of a modal feature extraction module, a dynamic gating fusion module and a classifier through the meta-learning model, the dynamic adaptation module comprises a meta-learning model constructed based on an MAML framework, and is used for optimizing global parameters of each module according to vulnerability features learned in pre-training; and inputting the multi-modal fusion feature vector into a classifier, and generating and outputting a vulnerability detection result of the smart contract. According to the method, known vulnerabilities can be accurately detected by fusing multi-modal features, and novel vulnerabilities can be rapidly adapted and detected.

The invention discloses a large reasoning model factuality enhancement method, system and device based on reinforcement learning and a medium. The method comprises the following steps: constructing a large language model reasoning thinking chain factuality detection and enhancement framework (REELANCE), and performing fine-grained fact accuracy evaluation on a reasoning chain output by a large language model; based on the result of the fact check classifier, the reasoning chain factuality is enhanced by adopting a reinforcement learning group strategy optimization technology (GRPO); a multi-dimensional reward mechanism is designed to improve the fact accuracy of the reasoning step; quantitatively analyzing the influence of enhanced training on the nerve activation change in the model through a mechanism interpretability technology; a comprehensive optimization target giving consideration to the fact accuracy and the reasoning quality is established, the fact robustness is improved, the standard reasoning performance is kept, and the balance between the fact and the reasoning ability is achieved; the key problem of inaccurate facts of an inference type large language model in a high-risk scene can be effectively solved; the invention also provides a system, equipment and a medium for realizing the method.

Hierarchical modelling and advanced feature engineering discover abnormalities in time series data with irregular trends. Data is collected in real time to ensure temporal integrity in the invention. Extraction filters and isolates useful data. Data cleansing removes noise and extraneous data after preliminary analysis identifies patterns and abnormalities. Feature engineering organizes cleansed data for machine learning algorithms. Primary storage stores this data for fast retrieval and extensive trend analysis. Holidays and weekends provide unique patterns in trend analysis. These trends are used to cluster data and create hierarchical predictive models, starting with a first-order model for general trends and increasing in order to refine residuals. Serializing these models improves storage and retrieval. Trend clusters are created from new data points, and algorithms detect pattern deviations. Statistical tests and machine learning classifiers identify anomalies and create alerts and remedial measures. The system monitors and analyzes incoming data to detect anomalies.

The invention discloses an incomplete multi-view multi-label data classification method based on semantic enhancement and pseudo-label uncertainty perception, and the method comprises the steps: employing a dual-channel feature extraction and decoupling module to obtain the shared semantic representation and specific representation of each view in each sample for a constructed incomplete multi-view multi-label data classification network model; performing cross-view fusion on the shared semantic characterization and the specific characterization, obtaining a unified shared characterization and a unified specific characterization corresponding to each sample, performing feature fusion, obtaining a fusion characterization of each sample, inputting the fusion characterization of the sample output by the dual-channel feature extraction and decoupling module into a classifier for multi-label prediction, and performing multi-label prediction on the fusion characterization of the sample. Therefore, a multi-label classification prediction result is obtained, and model training is carried out based on a total contrast learning loss function and a joint supervision classification loss function. According to the method, the classification performance and the model robustness on incomplete multi-view multi-label data are remarkably improved through training learning under the guidance of semantic enhancement and uncertainty.

The invention relates to a food reserved sample quality monitoring method based on machine learning, and the method specifically comprises the following steps: deploying a sensor array in a food reserved sample environment, collecting reserved sample multi-dimensional physicochemical characteristic data in real time, constructing a training set through historical normal and degraded batch data, and marking a state; the method comprises the following steps: constructing a machine learning model for food reserved sample quality monitoring, inputting sample data in a training set into the model, sequentially passing through a dynamic distribution alignment module, a physical constraint confrontation enhancement module and a multi-scale residual space-time network, and performing dynamic attention mechanism enhancement by utilizing physical constraint and environment modulation. And finally, outputting a quality category probability by the multi-modal classifier. Then calculating model loss, and carrying out iterative training on the model to obtain a trained model; and deploying the trained model at a reserved sample monitoring terminal, inputting newly collected monitoring data, and predicting the quality state of the food reserved sample. According to the invention, high-efficiency and real-time quality monitoring on the quality of the reserved food sample can be realized.

The invention discloses a ship intelligent fault diagnosis method and system based on open label space identification. The method comprises the following steps: obtaining a multi-source sensor time sequence signal of a ship system and constructing a training sample set; carrying out feature extraction on the training sample set by utilizing a deep neural network model, and strengthening the clustering characteristics of the features by adopting a center loss function in the training process; training a classifier on the basis of feature extraction and introducing an open set loss function to form a comprehensive objective function; extracting features from a to-be-diagnosed sample, embedding the features, calculating the distance between the to-be-diagnosed sample and a known fault category feature center, and judging an unknown fault through comparison between the minimum distance and a preset threshold value; and outputting a known fault category or triggering an unknown fault alarm according to a judgment result, and dynamically expanding and updating a model knowledge base based on accumulated unknown fault samples. The method can break through the limitation of the traditional closed set hypothesis, effectively identifies the unknown fault type, and achieves the self-adaptive learning and continuous optimization of a ship fault diagnosis system.

The invention discloses a federal learning method and system based on global prototype guidance. The method comprises the following steps: a server sends a global model and a global prototype to a local client; the local client performs rebalance comparison learning by using the global prototype, and updates the local model; the local client generates a balanced confrontation positive prototype instance set and trains a classifier; the local client side calculates a local prototype and uploads the local prototype and the updated local model to the server side; and the service aggregates the updated local model and local prototype, and updates the global model and global prototype. The system comprises a server and a local client. According to the method, the Non-IID problem in federated learning is solved, and the accuracy and generalization of the model are improved. The method can be widely applied to the technical field of distributed machine learning.

The invention discloses a vulnerability detection method and system based on abstract syntax tree decomposition and large language model assistance. The vulnerability detection method and system are used for solving the problem that an existing pre-training model is insufficient in detection accuracy under complex code logic and multiple execution paths. The method comprises the following steps: firstly, analyzing a code snippet into an abstract syntax tree, splitting the abstract syntax tree into a plurality of sub-trees through an improved decomposition algorithm, and combining each sub-tree with a natural language annotation generated by a large language model to form an abstract sub-tree with the annotation; then, a semantic aggregator based on Transform is used for modeling the relation between the sub-trees, features are fused to a target vulnerability vector, and finally, vulnerabilities are predicted through a classifier. Based on the technical scheme, the vulnerability detection accuracy is effectively improved, and the performance of the vulnerability detection model is greatly improved.

An adaptive classification retrieval-augmented generation model system and a method, relating to the technical field of artificial intelligence. The system comprises: a query splitting module, a classifier module, a processing module, a user interaction module, and an evaluation and optimization module. The user interaction module is used to receive an original query of a user, and display a generated answer. The query splitting module is used to split the original query received by the user interaction module into multiple equivalent queries. By means of integrating a classifier module, the complexities of different tasks are effectively identified and, corresponding processing methods therefor are determined, so that the understanding and application efficiency of the augmented generation technology is optimized. Advanced RAG technologies and specially adjusted classification algorithms are combined, so that not only is the accuracy of question answering improved, but processing speed is also improved, and consumption of running resources is significantly reduced.

An analysis engine receives data characterizing a multimodal prompt for ingestion by a generative artificial intelligence (GenAI) model. The multimodal prompt is processed and fed into a plurality of layers from which an intermediate result of the GenAI model or a proxy of the GenAI model is obtained. The analysis engine, using a prompt injection classifier and the intermediate result, determines whether the prompt comprises or is indicative of malicious content or elicits malicious actions. Data characterizing the determination is provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.

The invention discloses a semi-supervised text data multi-label classification method, system and device and a storage medium. According to the method, a double-branch model of a shared feature processing network is constructed, a pseudo label generator is utilized to automatically generate a pseudo label for an unlabeled sample on the basis of limited labeled data, and the pseudo label generator and a classifier are jointly trained to realize collaborative learning of labeled data and unlabeled data; by introducing an adaptive threshold mechanism and an improved loss function design, the recognition precision of minority class labels is effectively improved. Compared with a traditional full supervision model, the method has the advantages that the dependence on large-scale manual annotation is reduced, the data preparation cost is remarkably reduced, and the application performance of the classification model in multi-label scenes such as medical text analysis, public opinion monitoring and personalized recommendation is improved. The system, the device and the storage medium provided by the invention can realize the method, and have good expandability and engineering application value.

The invention provides a large language model generation code detection method and system, which is applied to the technical field of artificial intelligence, and comprises the following steps: obtaining a to-be-detected code; a to-be-detected code is input to a trained shared encoder, a code feature vector is obtained, the shared encoder is obtained through multi-target joint training, and the multi-target joint training is used for optimizing classification loss, domain confrontation loss, comparison loss and difficult sample loss at the same time; l2 normalization is carried out on the code feature vector, and the code feature vector is mapped to a hyperspherical space to obtain spherical embedding; the sphere is embedded and input into a sphere category classifier based on sphere logistic regression for classification processing, a detection result of the to-be-detected code output by the sphere category classifier is obtained, the detection result comprises AI generation and human writing, and a decision boundary of the sphere category classifier is an intersection line of a hyperplane and a hypersphere for classification. According to the invention, the AI generation code and the human compiled code can be accurately distinguished.

The invention discloses a voice emotion recognition method based on multiple scales and multiple features, and belongs to the technical field of artificial intelligence. The method comprises the following steps: firstly, preprocessing an audio signal and extracting a spectrogram and a Mel-frequency cepstral coefficient; then, a residual network, a bidirectional long-short-term memory network and a HuBERT pre-training model are respectively utilized to extract spectrogram high-order spatial features, time sequence context features and voice semantic embedding features; secondly, inputting the first two features into a multi-dimensional multi-scale feature extraction module to extract richer time-frequency features, performing deep fusion by using a multi-layer cross attention mechanism, and performing weighted fusion with speech semantic embedded features; and finally, all the advanced features are spliced, and a final emotion category is recognized through a full-connection classifier. According to the invention, through combination of multi-scale feature extraction and an advanced fusion mechanism, the problem of insufficient complex emotion modeling ability in the prior art is effectively overcome, and the accuracy and robustness of voice emotion recognition are significantly improved.

In some embodiments, a method for classifying elements of a blood sample is provided, the method including: digitally staining an image of the blood sample using a trained machine learning model to generate a digitally stained image; extracting one or more intermediate features generated by the trained machine learning model during the digital staining of the image; providing the one or more extracted intermediate features to a trained multi-class classifier; and employing the trained multi-class classifier to classify at least one element in the blood sample based on the one or more extracted intermediate features. Many other embodiments are also provided.

The invention discloses a steel structure construction safety risk dynamic assessment method and system based on artificial intelligence. The method comprises the steps of obtaining a multi-dimensional data set in real time; based on the multi-dimensional data set, a deep learning algorithm is adopted to fuse an image and a sensor signal, a structure defect is accurately identified, and a defect feature vector is extracted; when the defect feature vector exceeds a preset threshold value, a machine learning classifier is introduced to construct a risk assessment model, classification quantification of defects is realized, and an initial risk level is output; for the initial risk level, analyzing the space-time fusion data by using a probability state transition model, simulating the dynamic evolution trend of the risk and judging the stability of the risk; a mapping model is constructed based on a virtual simulation technology, dynamic simulation deduction is carried out on risk changes, and a visual visualization report and a graded early warning signal level are generated. According to the invention, full-chain intelligent closed loop from defect perception to risk prospective early warning is realized, and the safety management and control level and accident prevention capability of a construction site are greatly improved.

The invention discloses a training method of an image classification model for small sample incremental learning. The training method comprises the following steps: acquiring a pre-training visual encoder comprising a plurality of converter layers and a first classifier, and constructing a learnable prompt vector group to carry out reference stage iterative training; constructing a structure alignment branch parallel to the first classifier, and carrying out joint training by taking approaching the first classification output as a target to obtain a prompt vector group of structure alignment; extracting image features by using a prompt vector, constructing prototype matrixes of various categories, constructing a trainable low-rank residual matrix as residual compensation, forming a second classifier by the sum of the two matrixes, and training the second classifier; and finally outputting a visual encoder containing a prompt vector group and a second classifier for image classification in a reasoning stage. According to the method, the expressive power of the pre-trained visual language model can be fully utilized, and through low-rank structure alignment and continuous residual adaptation, a new category is efficiently learned while old-class knowledge is stably kept.

The invention discloses a subject corpus labeling method and system based on classifier multi-round iteration, and belongs to the field of natural language processing and machine learning. The method comprises the steps that a multi-level subject classification system is constructed, multiple rounds of iterative reasoning are conducted on a target subject based on a Fasttext classifier and a seed set, positive and negative samples are dynamically generated in each round of iteration, and classification accuracy is gradually improved; performing multi-dimensional scoring on the positive samples by using the large model to screen high-quality positive samples, and extracting keywords to optimize classification boundaries; a subject problem is generated through a WebQA method, and a low-recall-rate subject corpus is retrieved and supplemented; training an information filter to identify, filter and reject low-quality contents such as advertisements and garbage; and finally, efficient and accurate subject labeling is realized through a classifier and filter series connection process. According to the method, the problems of low efficiency, large resource consumption and subject understanding deviation of traditional labeling are solved, and the method is suitable for efficient and accurate subject labeling scenes of large-scale text data.

The invention discloses a classification model privacy disclosure risk black box detection method and device, and relates to the technical field of machine learning security, and the method comprises the steps: constructing an auxiliary data set; generating an auxiliary model covering different privacy risk levels; training a shadow model and a black box member reasoning attack model corresponding to the shadow model; evaluating the privacy risk score of each auxiliary model by using the attack model, and labeling a high-risk / low-risk label; calculating a correction prediction entropy difference, and screening the first k samples with the maximum difference to form a query data set; extracting a prediction result of each auxiliary model on the query set, calculating and correcting a prediction entropy and a mean value, a variance, a kurtosis and a skewness thereof, and splicing into a k + 4-dimensional risk feature vector; training a risk detection classifier; similar features of the model to be detected are extracted and input into the classifier, and the classifier performs five-level privacy disclosure risk level mapping according to the final risk probability prediction value and outputs the five-level privacy disclosure risk level mapping. The method is mainly applied to risk assessment of classification models in high-privacy sensitive fields such as financial credit investigation and medical diagnosis.

The invention discloses a multi-modal sentiment analysis model based on hierarchical adaptive cross-modal fusion, belongs to the field of natural language processing, and is used for solving the problems that in the prior art, multi-granularity sentiment feature extraction is insufficient, a cross-modal fusion mechanism is rigid, and the distribution difference between different-source modals is large. The method comprises the following steps: firstly, extracting features of texts, audios and visual modalities from original video data, and coding the features into advanced semantic features; secondly, multi-granularity information is fused through a hierarchical feature extractor to generate enhanced single-mode features; then, a self-adaptive cross-modal fusion network with a text as a core is adopted to realize bidirectional interaction and dynamic weighted fusion between modals; further, a dynamic contrast learning mechanism is introduced to align modal distribution in a unified potential space; and finally, inputting the optimized multi-modal features into a classifier and outputting an emotion analysis result. According to the model, through collaborative optimization of multi-granularity feature extraction, adaptive fusion and comparative learning, the accuracy of sentiment analysis and the robustness of the model are remarkably improved.

Active machine learning systems for anomalous event detection and classification. Initial samples from an industrial environment may be received and labeled. Initially, a training pool of audio samples may be labeled. These labeled samples may be used to train an audio event classifier to detect and categorize sounds. Environment states may be calculated using outputs from the classifier. A batch of audio samples may then selected from an unlabeled pool for annotation, guided by a reinforcement learning agent. These selected samples may be annotated and added to the labeled training pool. The classifier may be retrained with this updated pool. Rewards may be calculated for each of the annotated samples based on their annotations. The environment states may be updated using the retrained classifier, and the exploration-exploitation parameter of the reinforcement learning agent may be adjusted. The reinforcement learning agent may be retrained using the updated environment states and rewards.

A method includes: receiving a query; generating, using a task classifier, a classification associated with the query; generating a score and a context for a subset of large language model providers, among a plurality of large language model providers, that provides a highest correlation with the classification associated with the query; determining, using a contextual bandit, a large language model provider, among the subset of large language model providers, based on a trained model for the contextual bandit; providing the query to the large language model provider; receiving a response from the large language model provider; and updating the trained model for the contextual bandit based on the response.

The invention discloses a numerical control machine tool part machining data storage method and system. The numerical control machine tool part machining data storage method comprises the steps that multi-source data in the machining process is collected and synchronized in real time; performing process-entropy-based process section segmentation on the data stream, extracting multi-modal features, evaluating data values through a classifier fusing rules and machine learning, and generating feature data packets according to a hierarchical storage strategy; a data gene code is bound for each part, a processing process knowledge graph is constructed, and full-chain data association and accurate tracing are achieved; a sample is generated based on the feature data packet, a graph neural network and Bayesian optimization are utilized to train a process optimization model at a cloud end, optimization parameters are fed back to a machine tool to be executed, and a closed loop is formed. Process entropy segmentation, knowledge graph and GNN closed loop optimization are adopted, the problems that data storage cost is high, tracing is difficult, and data values are not mined are solved, and the accuracy of data tracing is improved. And intelligent data compression, full-life-cycle tracing and process adaptive optimization are realized.

A system for dynamically adapting a conversational artificial intelligence (AI) system includes a chatbot system, a feedback and classifier unit, and a document generator. The chatbot system generates a response to a user query. The feedback and classifier unit receives the user query, a large language model (LLM) provided response, and system architect provided feedback to create a data object. The unit retrieves a set of ternary questions from a questions database and processes the data object using the LLM to generate answers, creating a feature vector of ternary answers. It then determines a classification label for the feedback by processing the feature vector with a decision tree, where the label indicates a knowledge or behavioral update. The document generator creates a new document based on the classification label and feedback and updates a knowledge base or prompts database with the new document based on the determined classification label.

A system and method for dynamically optimizing large language model (LLM) inference by selectively deactivating layers based on query complexity. A multi-label classifier is trained on diverse user queries and their optimal layer configurations. During inference, the classifier analyzes incoming queries to predict which LLM layers can be safely deactivated without compromising output quality. The system processes user queries through the LLM with the predicted layer configuration, reducing computational resources while maintaining accuracy. A database stores historical queries, layer configurations, and performance metrics for continuous system improvement.

A computer-implemented method comprises providing images and attributes of a damaged vehicle that has been damaged in a collision event to a trained computer vision machine learning model, which in response provides indicators of physical damage sustained by the damaged vehicle during the collision event; providing the indicators to a trained classifier machine learning model, which in response provides a predicted class of bodily injury sustained by an occupant of the damaged vehicle during the collision event and a confidence indicator representing a level of confidence that the predicted class of bodily injury is correct; and providing the predicted class of bodily injury and the confidence indicator to an analyst for use in evaluating a bodily injury claim related to the occupant of the damaged vehicle and the collision event.

The invention discloses a power frequency waveform classification method and device, a medium and equipment, and belongs to the field of waveform classification, and the method comprises the steps: obtaining preprocessed power frequency waveform data of a power transmission line, calculating various waveform features, and inputting the features into a preset cascade classifier model. The model is composed of a plurality of XGBoost binary classifiers which are cascaded in sequence, and each classifier is responsible for executing a specific binary classification task. The output result of the previous classifier is used for triggering the starting of the next classifier, or the final classification result is directly output. The cascade structure not only improves the accuracy and efficiency of classification, but also reduces the complexity of the model, and effectively solves the problem that the fault positioning power frequency of the power transmission line cannot be accurately and efficiently classified in the prior art.

The invention belongs to the technical field of GIS fault detection, and discloses a GIS latent insulation fault diagnosis method and system under strong noise, and the method achieves the omnibearing signal capture through synchronously obtaining optical, mechanical, electrical and acoustic signals, and solves a problem of weak signal leak detection. For strong noise interference, noise and fault characteristic frequency bands are effectively separated by using multi-modal signal difference and constructing a time domain and wavelet domain cascaded double-layer dictionary. A discriminative sparse model is combined with a K-SVD algorithm to carry out end-to-end iterative optimization, sparse features with higher discriminative force are automatically mined, and the problem of deep feature mining is solved. For the problem of weak generalization ability of small samples, a classifier error term is introduced into an objective function, a joint optimization objective function fusing reconstruction and classification errors is constructed, dictionary learning and classifier training are combined into one, the generalization ability of the model is significantly enhanced, and the method is suitable for large-scale popularization and application. Therefore, timely and accurate identification of the latent insulation fault under the conditions of strong noise and small samples is realized.

The invention is suitable for the technical field of intelligent traffic systems, and provides a trajectory anomaly detection method based on structure enhancement and comparison pre-training. The method comprises the following steps: firstly, carrying out road network matching on an original GPS track, extracting attributes such as a road type and a road length, coding the attributes into a structure lexical element sequence, and constructing a structured Prompt text; then, through structure-semantic comparison pre-training, enabling the model to learn an alignment relationship between path structure features and text description thereof, and obtaining stable shared representation; in the fine tuning and reasoning stage, the aligned path structure vectors are used for carrying out structure guiding fusion on large language model coding representation, track behavior categories are output through a classifier, and meanwhile key evidence fragments are given according to attention weights so as to enhance interpretability. Compared with an existing method, the method has the advantages that road structure information is explicitly fused, cross-modal alignment is achieved, and the accuracy and interpretability of track anomaly detection and the generalization ability in a city-wide cross-regional scene are remarkably improved.

The invention discloses a traffic accident detection method based on a VAE model, and the method comprises the following steps: obtaining original data, and carrying out the primary processing of the obtained data; dividing a data set; converting a data format into a tensor format; performing cross validation on the data; pre-training the VAE model, adjusting and training a classifier, calculating a sample reconstruction error, KL divergence, a potential spatial distance and an output probability of the classifier, and calculating a mixed score through a mixed scoring formula; according to the method, the accuracy of a traffic accident detection model is improved, the accuracy of accident detection in an unbalanced data scene is effectively improved by fusing the characterization learning ability of the VAE and the discrimination ability of the classifier, and a mixed scoring mechanism is further introduced, so that the accuracy of the traffic accident detection in the unbalanced data scene is improved. Information of multiple dimensions such as VAE reconstruction error, KL divergence, potential spatial distance and classifier output probability is integrated, and the robustness of the model is enhanced.