1151 results about "Confidence threshold" patented technology

The present disclosure relates to the technical field of target detection, and particularly relates to a model deployment method, an end-side device and a storage medium, which are used for solving the problem in the related art of the accuracy of a deployed model being low. The method comprises: performing target detection on a video frame image input into a first model, and acquiring a first target detection result and a first confidence; if the first confidence is greater than or equal to a first confidence threshold value, recording the video frame image and the first target detection result as samples in a training set; if the first confidence is less than the first confidence threshold value, performing target detection on the video frame image on the basis of a second model, and recording the video frame image and an acquired second target detection result as samples in the training set; and training the first model on the basis of the training set, and replacing the current first model with a trained first model for subsequent target detection. In this way, the accuracy and model generalization capability of a first model are improved.

An AI-driven financial planning system for real-time market adjustment, consisting of: a neural inference coprocessor configured to execute deep financial forecasting models, including recurrent neural networks and attention-based encoders, on the device, and wherein the processor dynamically updates portfolio parameters in response to market signals exhibiting volatility differences above a statistical threshold calculated using an exponentially weighted moving standard deviation; a financial data acquisition module configured to continuously receive and analyze heterogeneous data streams, including market indices, interest rates, stock and bond price fluctuations, economic indicators, regulatory updates, and financial news sentiment feeds; a behavioral analytics engine configured to create a dynamically evolving user-specific financial behavior profile based on real-time analysis of transaction history, income-expenditure cycles, psychometric test results, and temporal lifestyle patterns using supervised and unsupervised machine learning algorithms; A goal optimization module configured to transform high-level, user-defined financial goals into quantitatively tracked multi-level goals. It uses a reinforcement learning framework that predicts optimal asset allocations across multiple time horizons. a real-time strategy simulation engine configured to perform Monte Carlo simulations and deep Q-learning-based assessments to simulate the resilience of proposed financial strategies under different macroeconomic regimes and trigger redistribution events based on predefined confidence thresholds; a compliance-aware execution interface configured to interact with financial institutions through encrypted API channels, ensuring policy enforcement using a smart contract validator and a hardware-enabled secure transaction signing unit; and a recommendation display unit configured to render dynamic dashboards for visualizing investments, reallocation warnings, confidence intervals, and sensitivity sliders, and where user interaction with the unit flows back into the behavioral model for real-time learning.

The invention relates to the technical field of text classification, and provides a text classification method based on a large model and a rule engine, and the method comprises the steps: S1, storing multi-level rule classification labels, and constructing a classification rule template library; s2, receiving text data from various data sources, and preprocessing the text data; s3, performing rule matching on the text data based on the classification rule through a rule engine, and outputting a rule classification result; and S4, when any one of the following conditions is met, large language model classification is triggered: a, a classification rule is not matched; b, matching a classification rule, wherein the rule confidence is smaller than a rule confidence threshold; c, the text data length exceeds the preset text data length; d, matching a specific business scene label; outputting a model classification result; and S5, when the rule engine classification in the S3 and the large language model classification in the S4 are parallel, executing the strategy. The output reliability and the service continuity are guaranteed, and the method is suitable for scenes with high accuracy requirements such as financial compliance examination and the like.

The invention provides a thin film surface defect detection method and system, and relates to the technical field of defect detection.According to the thin film surface defect detection method and system, an intelligent secondary verification link is constructed by introducing a defect confidence evaluation mechanism based on form and energy distribution, so that the detection performance is fundamentally improved; according to the mechanism, real physical defects with regular forms and concentrated energy and pseudo defects caused by electromagnetic interference, instantaneous film wrinkles and the like can be accurately distinguished, and the problem of high false alarm rate caused by dependence on single signal strength in the prior art is effectively solved while the high detection rate of low-contrast defects is reserved; besides, the judgment model based on physical characteristics has natural robustness for background noise generated in high-speed motion, and an adjustable confidence threshold value endows the system with extremely high practical flexibility, so that the system can adapt to complex and changeable industrial environments and different quality control standards, and the method is suitable for large-scale popularization and application. And the accuracy, the reliability and the intelligent level of the whole detection system are obviously enhanced.

An apparatus and method for static image of time series measured data to time series translation is disclosed. The apparatus comprises at least a processor configured to receive a static image of time series measured data, convert that static image from its initial domain to a usable time series within another user-selected domain, then to validate the conversion against a confidence threshold.

The invention relates to the technical field of voice interaction, particularly discloses an AI intelligent robot voice interaction method and system, and aims to solve the problems of low voice interaction accuracy, insufficient reliability and lack of authority control in a complex noise environment. A dynamic noise feature library containing steady-state noise, impact noise and human voice interference features and a pre-stored gesture instruction library are constructed, audio signals are collected in real time, low-frequency-band, middle-frequency-band and high-frequency-band differential noise reduction is executed, Mel-frequency cepstral coefficient features are extracted, noise scenes are matched, corresponding voice recognition models are switched, and voice recognition is achieved. And calculating a confidence value of the voice instruction, outputting multi-modal verification data in combination with a dynamic confidence threshold, and outputting an authority control signal through voiceprint matching, authority verification and instruction consistency judgment. Through multi-modal fusion, dynamic adaptation and authority control, the voice recognition accuracy and interaction safety in a complex noise environment are remarkably improved, and the method is suitable for scenes such as factory intelligent inspection.

A generative mediator engine can perform a requested interaction with a conversational agent of a target entity on behalf of a user. An internal conversational platform can identify intents for the requested interaction. An external artificial intelligence engine can perform intent discovery when an intent is not identified above a confidence threshold. A discovered intent unknown to the generative mediator engine can be received from the external artificial intelligence engine and used, with input requirements determined by the generative mediator for the requested interaction, by a dialog generator to generate a sample dialog for the requested interaction. User feedback can be received after review of action items and expected inputs identified from the sample dialog. The generative mediator engine can perform the requested interaction with the conversational agent on behalf of the user and without receiving user intervention during the requested interaction.

The invention relates to a self-healing operation and maintenance method and device for a big data component, equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing and feature fusion of multi-modal data of the big data component, obtaining and inputting a high-dimensional component health state vector into a fault diagnosis large model, and outputting a target fault type and a root cause confidence coefficient; when the root cause confidence coefficient meets a preset confidence coefficient threshold value, determining a historical restoration strategy based on the target fault type, calculating cluster environment similarity based on the cluster environment characteristic parameters of the current cluster environment and the historical restoration strategy, and if the cluster environment similarity meets a preset cluster environment similarity threshold value, determining that the current cluster environment is abnormal. If yes, taking the historical restoration strategy as a candidate restoration strategy; if not, a candidate repair strategy is generated based on a reinforcement learning strategy generation engine, and the candidate repair strategy is executed based on the strategy level matched with the candidate repair strategy; compared with the prior art, the technical scheme of the invention can effectively solve the problems of leak detection, low efficiency, solidification and the like of traditional operation and maintenance.

The invention relates to an audio data processing method and device, electronic equipment and a vehicle, and is applied to the technical field of computers. The audio data processing method comprises the following steps: acquiring target voice data of a user; obtaining at least one grammar element corresponding to the target voice data, and determining a grammar structure integrity score corresponding to the target voice data based on the at least one grammar element; determining an environment interference score corresponding to the target voice data; determining a confidence score corresponding to the target voice data based on the grammatical structure integrity score and the environmental interference score; and under the condition that the confidence score is greater than or equal to the target confidence threshold, determining to execute a target control instruction corresponding to the target voice data, thereby determining whether the target voice data is an instruction for vehicle control by combining the grammatical structure integrity of the voice data and the environmental interference degree. The accuracy of vehicle control and the interaction experience of the user and the vehicle are improved.

The invention discloses an image recognition method and system based on multi-modal data fusion, and the method comprises the steps: synchronously collecting image data and non-image modal data for a target scene, carrying out the preprocessing of the image data and the non-image modal data, inputting the preprocessed data into a feature fusion network, and carrying out the recognition of the feature fusion network; layered visual features of image data are extracted through image branches of the feature fusion network, modal features of non-image modal data are extracted through modal branches, the extracted features are fused through an interaction layer, and a fusion feature vector containing cross-modal associated information is formed; and inputting the fusion feature vector into a classifier to identify information and confidence of a collection object in the target scene, and when the confidence is lower than a confidence threshold, performing feature extraction and recognition again to form collection-fusion-recognition-feedback closed-loop optimization. In this way, the internal relation and complementarity between different modal data are fully considered, so that the accuracy of image recognition is improved, and the adaptability of the recognition process is enhanced.

The invention provides a commodity identification method and device, and the method comprises the steps: carrying out the multi-view collection and preprocessing, and constructing a second color image set and a depth image set; performing three-dimensional reconstruction and volume estimation based on the second color image set and the depth image set, and judging a commodity packaging form; extracting a commodity packaging semantic and price information area based on the second color image set, constructing a space-semantic joint code, and generating global packaging semantic features; based on the second color image set, a multi-view visual coding network is adopted, and a visual prediction SKU index is output; carrying out structure-semantic fusion modeling based on multi-source heterogeneous features, and outputting commodity category labels; and matching a commodity price based on the commodity category label, generating an interaction prompt sign in combination with a confidence threshold, and triggering interaction feedback. According to the invention, commodity vision, space and price activity information are fused, the identification accuracy and price verification intelligence are improved, and the method is suitable for the field of intelligent cashier and retail automation.

The invention relates to the field of electric power material warehouse management and control, in particular to an electric power material digital intelligent professional warehouse management and control method, which comprises the following steps: acquiring electromagnetic spectrum data, identifying a central frequency point energy distribution diagram and an electromagnetic interference intensity level, constructing a three-dimensional electromagnetic field intensity distribution model of a professional warehouse, and dynamically adjusting the working mode of an RFID reader-writer; the method comprises the following steps: deploying a multi-modal check node in a storage area of a professional warehouse, and generating a composite check result through an infrared contour matching method and a UWB coordinate verification method when the confidence coefficient of electric power material data read by an RFID reader-writer is lower than a preset confidence coefficient threshold value; updating an electric power material inventory database according to a composite verification result, and performing electric power material inventory state analysis in real time through an edge computing node; the problems that inventory data distortion and material positioning deviation are caused by reduction of the accuracy rate of RFID in a strong electromagnetic environment, and inventory accuracy management of an electric power material warehouse is seriously restricted are solved.

The invention discloses a multi-modal perception and optimization method and system for a low-power-consumption AR device, and the method comprises the steps: collecting multi-modal data, task demands, resource state data and environment data for the AR device; lightweight processing is carried out on the multi-modal neural network model through model pruning, parameter quantification and distillation technologies; inputting the collected data into a lightweight multi-modal neural network model for dynamic reasoning to obtain a multi-modal recognition result; comprising the steps of executing modal adaptive weight acquisition based on task requirements and environment data; executing energy consumption constraint scheduling according to the equipment resource state data, dynamically selecting a reasoning path strategy, and obtaining corresponding modal feature output; multi-modal feature fusion is carried out, task reasoning is completed, and a multi-modal recognition result is obtained; early-leaving control is executed based on a middle-layer confidence coefficient threshold value in the reasoning process; and interactively outputting a real-time multi-mode identification result. According to the invention, energy efficiency and precision balance and multi-mode fusion low-power-consumption optimization can be realized.

The invention relates to the technical field of artificial intelligence and material engineering, and discloses a formula optimization method and device, equipment and a storage medium, and the method comprises the steps: carrying out the knowledge extraction of an obtained structured formula data set and unstructured technical literature data in response to a formula optimization target input by a user, and generating a table literature knowledge set; inputting the table literature knowledge set and the formula optimization target into a large language model to obtain a generated text corresponding to the formula optimization target; performing logic rule screening and risk assessment on the generated text through a knowledge fusion layer to obtain text output conforming to a confidence threshold, and generating a new formula scheme according to the text output; and performing multi-objective optimization on the new formula scheme based on a preset experimental cost constraint condition, and outputting an optimized recommended formula and a support evidence chain. By automatically fusing innovative components and process information in literatures, the output recommended formula has scientific basis and interpretability, and the practicability and innovativeness of an automatic formula are improved.

The invention relates to the field of visual processing, and provides a visual classification processing method and device based on large model and multi-modal data fusion. The method comprises the following steps: inputting a to-be-classified input image and a corresponding category text description into a text encoder for multi-level feature extraction to obtain global text features and local text features; performing fine-grained cross-modal alignment on the local visual features and the local text features, calculating association weights between the image regions and the text phrases through a bidirectional cross attention mechanism, and generating aligned intermediate features; splicing and fusing the aligned middle features and the global visual features, and inhibiting background noise in a fusion result and reinforcing discriminative features in the fusion result through a feature mask algorithm in combination with the global text features to obtain multi-modal fusion features; and synchronously inputting the multi-modal fusion features into a multi-space classifier to generate respective classification results, and adaptively outputting an image classification result according to a confidence threshold in combination with a dynamic routing mechanism.

The invention relates to the technical field of artificial intelligence and medical decision systems, and discloses a reasoning agent construction method and system for intelligent medical treatment guidance, and the method comprises the steps: constructing a hierarchical modal completion network and a modal correlation knowledge graph; constructing a medical feature cross-modal mapping network based on comparative learning, and mapping different modal medical data to a shared feature space; according to the uncertainty of the complemented data, constructing an uncertainty quantitative model and automatically adjusting a diagnosis confidence threshold; aiming at different disease types and symptom combinations, constructing a disease modal incidence matrix and a modal reliability evaluation network; through weighted voting, evidence convergence and specialist authority evaluation, cross validation and collaborative decision-making of multiple specialist knowledge are realized, and a final diagnosis suggestion is formed; according to the invention, the problem of limited diagnosis capability caused by lack of medical data in a medical resource limited environment is solved.

Artificially intelligent systems and methods for financial coaching provide personalized, fiduciary-compliant financial guidance through advanced machine learning architectures with measurable performance criteria. The systems implement privacy-preserving processing pipelines that detect personally identifiable information using multi-layered pattern recognition including regular expressions for formatted data sequences, named entity recognition with confidence thresholds above 0.85, and contextual analysis algorithms. A multi-step artificial intelligence processing workflow includes automated language detection, emotional tone classification with confidence scoring, financial profile transformation using predefined templates, context-aware question rephrasing, and semantic similarity matching employing vector embeddings with financial domain vocabulary weighting applying multiplier values between 1.3-2.0. Specialized training methodologies expand datasets through mathematical transformation functions utilizing statistical standard deviations with incremental variations between 0.5-2.0. Mood-based escalation logic automatically transfers users to human advisors when emotional indicators exceed confidence thresholds above 0.8. The systems maintain response times below 5 seconds while providing regulatory compliance through curated content sources and predefined fiduciary instruction parameters.

The invention discloses a pedestrian multi-target detection and tracking algorithm based on cross-layer fusion and dynamic adjustment, and aims at the problem that a deep convolutional neural network in a backbone network is frequently excessively parameterized, a composite scaling mechanism is introduced to reduce the parameter quantity and improve the feature extraction capability of the network; meanwhile, considering that targets with different scales exist in a data set, a lightweight cross-scale feature fusion module is fused at the neck of the network, so that the adaptability of the model to scale change is enhanced; besides, aiming at the problems of small targets and shielding targets, a new loss function InnerWiseWIoU is designed, and in combination with internal region optimization and context information, the target positioning precision is improved. According to the method, the new model is applied to pedestrian tracking, the shielding degree between pedestrians is calculated and the matching threshold is dynamically adjusted aiming at the condition that a fixed confidence threshold in a tracking task is not suitable for centralized and rapid change of a target, so that the tracking performance is effectively improved, and the adaptability in a complex scene is improved.

The invention relates to the technical field of transformer substation intelligent inspection, in particular to a meta-learning-based small-sample transformer substation equipment state adaptive inspection system, which comprises a state acquisition module for acquiring the current feature vector and environmental parameter data of a target node; the drift detection module is used for comparing environment parameters to judge data drift and dynamically adjusting a confidence coefficient threshold value; the risk assessment module inputs the feature data into a meta-learning model to output an initial risk probability, and generates an effective risk probability based on threshold filtering; the blind area measurement module is used for acquiring unobserved nodes and calculating system state blind area entropy; the scheduling decision-making module is used for comparing the blind area entropy with a threshold value and generating an entropy reduction bottom instruction or a self-adaptive routing inspection distribution instruction; the strategy updating module is used for extracting an actual inspection result and feeding back to the model for parameter updating; according to the invention, the scheduling difficulty when the resources are limited is solved, and the self-adaptive capability of the system under different environment interferences is improved.

The invention is suitable for the technical field of data processing, and provides a distributed data consistency control method and system, and the method comprises the steps: constructing an enhanced vector clock containing a fuzzy membership parameter and a time window deviation value; a dynamic fuzzy inference engine is built in combination with factors such as network delay jitter; calculating conflict event logic precedence relation probability distribution by utilizing an engine; generating an execution sequence hypothesis by integrating the multi-dimensional information; selecting an optimal scheme based on a dynamic confidence threshold; and resources and consistency are balanced through a three-level conflict processing mechanism. According to the method, the problem that a traditional vector clock cannot process conflict event sorting is solved, the accuracy and adaptability of distributed data consistency control are improved, and efficient collaboration of the system is guaranteed.

The invention relates to the technical field of deep learning, provides an optimization method of a defect detection model and defect detection equipment, and can be used for industrial quality inspection. According to the method, at least one to-be-detected product data set in production is used for carrying out multi-round iterative optimization on a defect detection model after pre-training, the defect that traditional model training is isolated from a production line environment is overcome, and closed-loop iterative detection and optimization of the model are achieved. In each optimization process, confidence coefficient learning and active learning are combined, the uncertainty of a model is quantified through confidence coefficient learning, a global confidence coefficient threshold value is determined according to the confidence coefficient corresponding to the defect type of each piece of to-be-detected product data, and the threshold value is dynamically adjusted by using the recall rate and the number of optimization times. And during active learning, sample data screening is carried out by using the adjusted global confidence threshold, so that blind labeling or over-fitting labeling is avoided, the quality of the screened samples is ensured, and the accuracy and generalization of a retraining defect detection model are improved.

The invention provides a multi-sensor fusion welding temperature monitoring method and a related device, and the method comprises the steps: dividing a welding region into a plurality of sub-regions according to the spatial distribution of the welding region, and obtaining the initial temperature data corresponding to each sub-region according to various temperature sensors; performing time alignment on the initial temperature data through difference resampling of different sampling points to generate a time sequence temperature matrix; performing real-time quality detection on the time sequence temperature matrix to generate a real-time quality score, and performing weighted combination on the real-time quality score and a preset confidence coefficient parameter table to obtain a dynamic confidence coefficient result set; fusing the effective temperature data exceeding a preset confidence threshold in the dynamic confidence result set according to a fusion rule corresponding to each sub-region to obtain a fused temperature result; and the fusion temperature results of the sub-regions are subjected to time sequence splicing to obtain a global temperature curve of the welding process, so that the accuracy of temperature detection in the welding process can be effectively improved.

The invention provides a BIM component automatic identification and storage method and system based on multi-modal artificial intelligence fusion, and relates to the technical field of building information, and the method comprises the steps: receiving a BIM component model file, and preprocessing an obtained attribute parameter table; inputting the information into a multi-modal recognition engine, and outputting a classification result and confidence through Rapidfuzz fuzzy matching file names, OPEN-CV + YOLOv8 image recognition and knowledge graph analysis parameters; the dynamic fusion decision-making module calculates the final confidence coefficient according to the dynamic weight, and solves classification conflicts in combination with a dynamic rule base of industry specification rule priority; according to a confidence coefficient threshold value, judging automatic warehousing or triggering grading manual auditing; manual audit data are collected and fed back to the multi-mode recognition engine, and incremental knowledge self-learning is achieved. Through multi-modal fusion and dynamic decision making, the BIM component recognition accuracy and warehousing efficiency are remarkably improved, personal errors are reduced, and dynamic updating of a component resource library and system self-evolution are achieved.

The invention discloses a cross-view image geographic positioning method based on dynamic threshold pseudo tag self-training learning, and the method specifically comprises the following steps: introducing a difficult sample feature mining method, dynamically adjusting the loss weight of a sample according to the change of similarity, and building a dynamic difficult sample triple loss model; the method comprises the following steps: dynamically adjusting a confidence threshold value of a sample by adopting an index moving average weighting method, iteratively training and screening an unlabeled sample, namely a pseudo label, establishing a pseudo label self-training mechanism of a dynamic threshold value, mining and utilizing non-paired data, and solving the problem of high manual labeling cost; a reference image most similar to a query image is found through image retrieval, and the offset of a query position is predicted. Experiments on CVUSA and CVACT data sets show that as the distance threshold increases, the accuracy of the cross-view image geographic positioning method based on dynamic threshold pseudo tag self-training learning presents a stable rising trend, and the cross-view image geographic positioning method based on dynamic threshold pseudo tag self-training learning is superior to other methods under the same threshold condition.

The invention provides a knowledge base and large language model-based product quality anomaly reason searching method, apparatus and device, and a medium. The method comprises the following steps: collecting quality anomaly-related historical multi-source heterogeneous data from a product life cycle-related system in real time or at regular time; the collected data are preprocessed; constructing a structured domain knowledge base; model training; performing abnormal feature extraction based on abnormal event triggering, and matching the extracted abnormal features with related causal association rules and causal atlas nodes in the constructed knowledge base to form a preliminary suspected reason set, and performing corresponding triggering condition conformity evaluation and historical case matching degree; calculating the final confidence of each candidate reason in the possible reason set through a preset fusion algorithm; at least one root cause is determined according to a set confidence threshold; and generating a targeted solution suggestion report based on the output root cause and the corresponding reasoning explanation text in combination with solution measures pre-stored in a knowledge base.

The invention provides a real-time data acquisition buffer throughput management method and device and electronic equipment, and relates to the field of data processing. Receiving a data object containing a sampling value, a timestamp, a measuring point number and a state mark through an input memory queue, and constructing a logic mapping identifier by a mapping agent unit to generate a logic write-in state; the logic write-in state is stored in a delay sensing buffer area, and a transaction confirmation delay sequence returned by a target database is monitored in combination with a distributed delay monitoring unit; establishing a time regression model based on the delay sequence, and generating a prediction projection window of transaction confirmation time; when the window meets a confidence threshold condition, releasing the logic write-in state to an output memory queue; and finally, writing into a database through an output thread, generating a completion mark table, and feeding back for delay buffer dynamic management. By implementing the technical scheme provided by the invention, the buffer data release rhythm and the database processing capability are conveniently coordinated, so that the overall throughput efficiency is improved.

The invention relates to the technical field of on-line monitoring, in particular to a flue gas cooler leakage monitoring system and method.The method comprises the steps that a multi-source real-time sensing layer module continuously collects tube plate weld joint area temperature data, cooling water flow characteristics and flue gas sulfur-containing gas concentration, and the sampling period is 250 milliseconds; the intelligent feature fusion module performs noise reduction processing and dynamic weight fusion on the multi-source data and outputs a leakage probability vector; the dual-system collaborative early warning module maps the probability vectors into confidence levels, a 65% confidence threshold triggers an acoustic verification system to start, an acoustic emission sensor array captures fractured sound wave features and then compares the fractured sound wave features with a voiceprint library through a convolutional neural network, and an early warning signal is output when the fractured sound wave features exceed a 0.85 similarity threshold; and the dynamic knowledge base updates the case characteristics online according to the early warning result. According to the system, second-level identification and continuously optimized closed-loop monitoring at the initial stage of leakage germination are realized, and the problem of response delay caused by intermittent detection in the prior art is solved.

The invention belongs to the technical field of semantic recognition, and discloses an extraction analysis method and system based on semantic expression understanding. The method comprises the following steps: performing semantic analysis and feature extraction on an original query statement input by a user in an interactive interface to generate a structured semantic unit; performing interaction scene judgment on the structured semantic unit based on a preset multi-dimensional judgment condition and then outputting a scene identifier; calling a corresponding target intention recognition strategy according to the scene identifier; inputting the structured semantic unit into a domain classification model, and outputting a plurality of candidate intentions and corresponding initial confidence; and determining a target user intention from the candidate intentions based on an interaction clarification result of the user and an intention query rule by using the initial confidence and a preset confidence threshold, and generating a standardized intention recognition result. According to the mode, the user semantics can be deeply understood, the recognition strategy is dynamically adjusted according to the context, and accurate, efficient and sustainable extraction analysis is carried out on the user intention through man-machine cooperation and closed-loop feedback.

Transactional Neural Reasoning AI (TNRAI) is a novel class of artificial intelligence designed to simulate human-like reasoning during live, multimodal user sessions. TNRAI departs from traditional static inference models by integrating a five-pillar architecture: (1) delta-path modeling for real-time outcome deviation detection, (2) skew-based adversarial recognition, (3) vector memory recall for behavioral context, (4) ambient reasoning overlays to incorporate situational data, and (5) a multi-logic arbitration engine that fuses rule-based, statistical, situational, and historical reasoning. The system evolves continuously via a CI / CD feedback loop, adjusting its logic and thresholds based on live outcomes. TNRAI supports overlays such as time or location constraints to influence reasoning and can activate conditional triggers based on logic patterns or confidence thresholds. This architecture enables adaptive, deliberative decision-making in real time, extending the utility of AI across domains such as automation, compliance enforcement, customer engagement, and transaction-based system control.

The invention provides an offshore ship suspicious navigation behavior identification method and system, and the method comprises the steps: collecting AIS, radar, satellite and other multi-source data of a ship in a target sea area, and carrying out the data cleaning, fusion and standardization preprocessing; extracting navigation features, staying features, mode features and interaction features of the ship, constructing a behavior recognition model based on a neural network, inputting ship features, and then outputting a recognition result and confidence; and judging result validity through a confidence coefficient threshold value of dynamic calculation, triggering artificial review or multi-source data secondary verification through an invalid result, generating early warning information when a valid suspicious behavior is identified, calculating a risk value in combination with a ship type, a behavior severity degree and an environmental impact factor, and matching a pre-constructed decision library to generate a decision suggestion. According to the method, the recognition precision is improved through multi-source data fusion and an intelligent model, and real-time and efficient monitoring and response to suspicious navigation behaviors are realized in combination with a confidence coefficient mechanism and risk decision support.