290 results about "Expert system" patented technology

The invention provides a shield intelligent auxiliary type selection system and method based on a large language model. The model selection system comprises a data input module, a rule knowledge base module, a large model reasoning module and a result generation module. The integrated decision-making system integrating a rule knowledge base, a deep learning model and expert system logic is constructed for the practical problems of complicated geological conditions, multiple rule constraints, high expert dependency and the like in shield construction, and the system combines a structured model selection rule and historical case data, and has the advantages of intelligence, standardization, self-learning, high efficiency and the like. The problems of low efficiency, high subjectivity, insufficient intelligent degree and the like of the existing shield tunneling machine model selection depending on artificial experience and partial standardized guide are solved, and the transformation of shield construction management from artificial experience to intelligent decision can be promoted.

The invention provides a bearing fault identification method based on a dynamic generative adversarial network and expert feedback, and relates to the field of bearing fault diagnosis, and the method comprises the steps: generating a high-fidelity fault vibration signal through employing a condition generator and a triple discriminator generative adversarial network; verifying and generating sample quality through a 1D residual verification network and adding the sample quality into a training set; segmenting the vibration signals passing the test by using layered adaptive sampling, and keeping high-frequency impact characteristics in the vibration signals; a dynamic sparse attention mechanism is adopted to reduce unnecessary attention calculation and improve calculation efficiency, and different types of faults are accurately recognized in combination with a hybrid expert system classifier; and detecting the confidence of the diagnosis result, and triggering a feedback mechanism to regenerate a sample to complete autonomous iterative optimization when the confidence is low. According to the method, a generative adversarial network, a fault diagnosis model and a feedback mechanism are fused, accurate diagnosis of bearing faults is achieved through multi-level data enhancement and screening feedback, the diagnosis precision is continuously improved in continuous iteration, and the method is suitable for solving the problem that a traditional method is poor in performance under data scarcity and noise interference. The innovative closed-loop evolutionary logic of generation-diagnosis-feedback is provided, and the robustness and accuracy of fault recognition are remarkably improved.

The invention discloses an unknown pollutant non-target identification method and system based on mass spectrum and a generative model, and is applied to the technical field of environmental monitoring and analytical chemistry. The method comprises the following steps: collecting mass spectrum data of a water body sample to be detected; preprocessing the original mass spectrum data, and outputting standardized features; inputting the standardized features into a pre-trained generative model to generate a plurality of candidate molecular formulas; using chemical and physical constraints to eliminate candidate molecular formulas which do not conform to rules; executing a rule-driven algorithm to obtain candidate pollutant molecular structures; carrying out comprehensive scoring and sorting on candidate pollutant molecular structures through chemical prior and environmental prior; and semi-quantitative or relatively quantitative concentration determination is carried out. According to the method, a rule-driven expert system and a data-driven generation model are combined, unknown pollutants which do not exist in a standard library can be effectively recognized and analyzed, and full-process automatic processing from original mass spectrum data to pollutant structures and concentrations is achieved.

The invention discloses a high-voltage power distribution station room dynamic environment and equipment state monitoring system and method. The system comprises a multi-dimensional data perception acquisition integration module, a heterogeneous data feature extraction and fusion transmission module, a pre-analysis module based on composite weight distribution, an association mapping module combined with historical data, a core diagnosis module based on an expert system rule base and a multi-modal visual output module. Multi-source data such as temperature and humidity, gas concentration, vibration, voltage and current are collected, after feature extraction and fusion, a data potential relation is mined by using composite weight distribution and historical data association, then equipment state evaluation and fault diagnosis are performed based on an expert system rule base, and finally, visual display and early warning output are performed in a multi-mode form. The system and the method can comprehensively monitor the environment and the equipment state of the power distribution station house, accurately diagnose faults, and effectively improve the operation safety and the operation and maintenance efficiency of the high-voltage power distribution station house.

The invention discloses an urban financial service full-link risk prevention and control management system, and belongs to the technical field of risk prevention and control. Comprising a multi-source heterogeneous data acquisition and treatment module, a dynamic risk portrait construction module, a real-time risk monitoring and early warning module, a risk disposal and decision support module, a cross-scene risk linkage analysis module, a compliance and privacy protection module and a system adaptive optimization module. Automation and manual cooperation of risk disposal are realized; a low-risk event automatically executes a preset action through a formula, so that the disposal efficiency is improved; medium and high risk events are pushed to an expert system, a disposal scheme is generated in combination with historical cases, and a block chain records an operation log to ensure transparency and traceability; the module balances machine decision and manual auditing boundaries, and the misjudgment risk is reduced.

The invention relates to an optimization method of a hybrid expert system, computer equipment, a readable storage medium and a program product. A plurality of experts contained in the MOE are deployed in a plurality of artificial intelligence chips in groups, and the method comprises the following steps: carrying out routing calculation on an original input tensor to obtain a routing calculation result; determining input element grouping information corresponding to each expert based on expert index information and expert weight information in the routing calculation result; taking the expert dimension as a parallel dimension, executing rearrangement operation for the original input tensor in parallel based on the input element grouping information, taking a rearrangement result as input information of an expert, executing matrix multiplication and accumulation operation, and taking the expert dimension as the parallel dimension, and based on the input element grouping information, executing anti-rearrangement operation of the matrix multiplication and accumulation operation result in parallel to obtain a final operation result. By adopting the method, the MOE reasoning performance can be improved.

The invention discloses an unsupervised industrial defect detection method, system and device based on a multi-mode expert system. The unsupervised industrial defect detection method comprises the following steps: acquiring a label-free industrial defect image data set and a text cue word; inputting into a multi-modal expert system to obtain a corresponding prediction result; screening out a preliminary labeling result; filtering the preliminary labeling result based on a prior constraint to obtain a fine label; inputting the industrial defect image data set and the fine label into a self-supervised model for training, and outputting a pseudo label after training is completed; performing double filtering on the pseudo labels by adopting a dynamic threshold value and a prior constraint; inputting the industrial defect image data set and the pseudo label after the current training double filtering into the self-supervised model after the last training for re-training, outputting the pseudo label, and returning to execute the double filtering until the trained self-supervised model is obtained; and predicting an industrial defect image to be detected by adopting the trained self-supervised model to obtain a corresponding industrial defect detection result. And the detection accuracy and efficiency are improved.

The invention discloses an incomplete multi-mode crisis event detection method based on a memory pool and a mode perception expert system, and relates to the field of machine learning, and the method specifically comprises the steps: building a prompt strongly related to an event instance based on the memory pool, and complementing a missing mode through a large model; a CNN encoder, a ResNet encoder and a CLIP encoder are utilized to construct a double-flow encoder, and text and visual feature extraction is achieved; capturing fine-grained event information and clues; the distribution difference between a complemented sample and a complete sample is identified based on a modal perception expert system, and the quality of fusion representation is improved; and inputting the mixed features into the classification head to complete a crisis event detection task. According to the invention, through the prompt construction module based on the memory pool, a large model is introduced to realize careful completion of a missing mode, and the distribution difference between a completion sample and a complete sample is accurately identified based on a mode perception expert system, so that a multi-mode data missing scene can be effectively coped with, and the accuracy and timeliness of crisis event detection are improved.

The invention discloses an automatic report generation method based on cross-modal agent collaboration, and relates to the technical field of automatic document generation, and the method comprises the steps that a central scheduling engine coordinates an outline generation agent, a data query agent, a chart drawing agent and a data analysis agent; the central scheduling engine drives the outline generation agent to generate a structured framework of a report based on a user demand, and dynamically schedules each agent according to a task dependency relationship; the data query agent realizes data retrieval by fusing retrieval enhancement generation and a thinking chain technology; the chart drawing agent adaptively generates a chart according to the data type; the data analysis agent is combined with an expert system and a knowledge base to generate an analysis conclusion; wherein the outline generation agent, the data query agent, the chart drawing agent and the data analysis agent are coordinated in parallel under the coordination of the central scheduling engine, and exception handling and dynamic adjustment are realized through a cross-agent coordination mechanism.

The invention discloses a network traffic anomaly detection method and system based on a multi-modal coupling Mamba model and a mixed expert. The method comprises the following steps: collecting network traffic; cleaning the network flow data, and standardizing the data format; shunting the cleaned network traffic data based on quintuple (a source IP, a destination IP, a source port, a destination port and a protocol type) information; extracting data packet length, transmission direction and load byte information, constructing a data packet length sequence and a load byte sequence based on a time sequence view angle and an interaction view angle, and constructing a layered data packet interaction graph; the multi-mode flow representation is input into a coupling Mamba model, deep coupling and dynamic interaction are carried out among different modes, and high-level feature representation with higher identification capacity is generated; dynamically selecting and activating a plurality of expert sub-networks through a gating network by utilizing a hybrid expert system; and a final flow detection result is generated in combination with a result output by the MoE classifier and the original features, and confidence feedback is provided.

A latent transformer architecture with latent attention mechanisms and expert processing systems for federated deep learning is disclosed. The system operates entirely within latent space, eliminating traditional embedding and positional encoding layers while maintaining full attention capabilities. Input data is compressed into latent vectors via variational autoencoder encoding, then processed by a latent attention module that computes query, key, and value matrices directly from latent representations. The architecture incorporates expert processing systems including gated latent expert networks for sparse computation and latent mixture of experts for collaborative processing. In the gated approach, a routing network selectively activates specialized expert modules based on latent vector characteristics. The mixture approach enables all experts to contribute through weighted combination, facilitating distributed computation and enhanced model expressiveness.

The invention discloses a fruit and vegetable disease and insect pest spectrum database dynamic evolution updating and diagnosis optimization method based on model confidence feedback driving. Comprising the following steps: acquiring spectral data; based on the disease diagnosis model, outputting a corresponding model confidence coefficient; executing a calibration module, and outputting a calibrated model confidence coefficient; based on a preset confidence model, judging whether the model confidence is not less than a threshold value of high model confidence; based on the expert system module, judging whether the corresponding model confidence is not less than a threshold value of low model confidence; if so, executing an expert reexamination process; if not, inputting an abnormal sample library; and obtaining the number of newly added expert samples, a low confidence ratio and an interval to the last training time, executing an incremental training module, and outputting an instruction for updating and optimizing the disease and pest spectrum database and the abnormal sample library. And a data updating and model optimization mechanism for cooperative work of experts and models is established, so that misjudgment accumulation caused by excessive confidence of the models is effectively prevented, and the reliability of entering a database is ensured.

The invention relates to a time sequence knowledge graph reasoning method fusing mixed expert and cross-branch contrast learning. The method comprises the following steps: inputting a historical time sequence knowledge graph into a trained time sequence knowledge graph reasoning model for time sequence knowledge graph reasoning; wherein the time sequence mapping knowledge domain inference model comprises a double-branch time sequence modeling architecture, a dynamic hybrid expert system, a cross-branch contrast learning module and a decoding scoring module; respectively capturing dynamic change in a short period and a long-term evolution rule of an entity in the whole time span through a double-branch time sequence modeling architecture; structural features are extracted through a dynamic hybrid expert system, and cross-timestamp relation dependency is mined; mutual information between local and global time sequence views is maximized through a cross-branch contrast learning module, and fusion and learning of different levels of time sequence features are promoted; and the decoding scoring module is used for time sequence knowledge graph reasoning. According to the method, the modeling precision of the sudden change event and the periodic change is remarkably improved.

Techniques for implementing a fuzzy logic expert system / service to control sending of a requested media file and / or control triggering of a remedial action (e.g., not sending the media file) are described. According to some examples, a computer-implemented method includes receiving a request at a content delivery service from a user for a media file, generating a piracy risk score for the media file by a fuzzy logic expert service of the content delivery service, sending the media file to the user based on the piracy risk score not exceeding a threshold, and blocking the sending of the media file to the user based on the piracy risk score exceeding the threshold.

The invention discloses a lightweight disease and insect pest early warning and capturing system, and the system comprises an execution unit which comprises an intelligent insect sticking plate, a trapping device, early warning equipment, and a controller; the acquisition unit comprises a multi-sensor array and an image acquisition device; the calculation unit comprises an edge processor, a data analyzer, an early warning calculation module, a configuration module and a decision module; the edge processor is configured with an image preprocessing method and a lightweight target detection and recognition algorithm; the early warning calculation module is configured with an insect pest counting algorithm, a density heat map generation method and a prediction model; the configuration module is used for configuring a threshold dynamic adjustment mechanism, an early warning grade division mechanism, an early warning trigger condition and a board change reminding mechanism; and the decision module is connected with the expert system. The technical problems of low manual inspection efficiency and limited coverage range are solved, the monitoring efficiency, the early warning precision and the prevention and control effect are improved, and the cost and risk of agricultural producers are reduced.

The invention relates to the field of big data analysis and artificial intelligence, in particular to a social public opinion perception and commodity demand prediction system and method based on multi-modal data, and the method comprises the steps: a multi-modal-text and image emotion fusion module extracts and processes text and image data through two-channel features, generates a text feature vector and an image feature vector, and carries out the feature extraction; a preprocessing module is used for fusing the multi-modal-text-graph emotion vector with the live broadcast activity data, the commodity historical sales volume and the live broadcast influence data to form a time sequence sensitive multi-modal-text-graph emotion fusion representation; the demand perception module performs adaptive feature conversion on fusion representation through a multi-layer perceptron to generate a multi-modal-text-graph fusion vector, and an expert system predicts commodity demands based on a decision theory in combination with sales data, label types and fan number information, so that effective extraction and bidirectional conversion of text and image modal features are realized.

The invention discloses a gear machine tool expert system construction method based on a generative rule, and the method comprises the steps: firstly carrying out the multi-dimensional modeling of a gear and a machine tool through an object-oriented technology, enabling a gear model to distinguish the characteristics of an axis position, a tooth curve and the like through an interface, and enabling a machine tool model to distinguish the machining technology and the precision grade through an interface; secondly, establishing a generation type rule mapping system from geometric, precision, material and functional attributes of the gear to machine tool movement, precision, process and rigidity attributes; finally, an intelligent reasoning system including forward reasoning, reverse verification, conflict coordination and self-learning iteration is constructed, and automatic and accurate matching from gear design requirements to machine tool optimization configuration is achieved. According to the method, the problems of high subjectivity and low efficiency caused by dependence on expert experience in the prior art are solved, and knowledge-driven intelligent decision making is realized.

The invention relates to the technical field of crossing of bioinformatics, artificial intelligence and plant genomics, and discloses a plant single cell gene expression prediction method, system, device and medium. A plant single cell gene expression prediction model realizes dynamic feature fusion of a DNA sequence and chromatin accessibility signals through a gated cross attention mechanism; the problem that a traditional single-mode model cannot model regulation and control dynamic association is effectively solved, and the result interpretability is enhanced; a hybrid expert system and a load balancing design are adopted to significantly improve the recognition capability of the model for rare cell types, and a decoupling prediction head design supports efficient transfer learning; a DNA long sequence processing mechanism and nucleosome scale feature coding ensure cross-species compatibility; an end-to-end automatic process and a dynamic parameter optimization framework greatly improve the practicability; a'prediction-verification 'closed-loop support system can be constructed for molecular breeding, and high-precision and interpretable prediction of plant single-cell gene expression is realized.

The invention relates to the technical field of desulfurization, in particular to a desulfurization slurry control method and device based on a self-attention mechanism, and the method comprises the steps: collecting parameters of a desulfurization system in real time, building a variable correlation thermodynamic map, and constructing a hybrid prediction model for simulating chemical reaction kinetic characteristics by using a delayed feedback mechanism; outputting a multi-target variable prediction result in the desulfurization efficiency, the SO2 concentration at an outlet and the pH value of the slurry; generating a slurry supply strategy based on the multi-target variable prediction result and a pre-constructed slurry supply strategy optimization model in combination with a preset expert system rule; and dynamically optimizing a slurry supply strategy according to the operation state of the desulfurization system. Therefore, the problems that in the prior art, due to the fact that univariate adjustment and empirical parameter setting are generally taken as the principal thing, the overall modeling capacity for the multivariate coupling relation in the desulfurization process is lacked, control response lag is likely to be caused, the adjustment precision is insufficient, robustness to external disturbance is weak, and emission stability and operation economical efficiency are difficult to consider at the same time are solved.

The invention relates to the technical field of stereo garage safety management, and discloses a garage parking space safety risk monitoring method comprising the following steps: S1, multi-dimensional information real-time acquisition: comprehensively acquiring vehicle state, equipment operation data, environmental parameters and other multi-dimensional information through sensors at three levels of vehicle, equipment and environment; basic data is provided for subsequent analysis; and S2, data fusion and dynamic calibration: carrying out fusion processing on the collected multi-dimensional data, and introducing a dynamic calibration mechanism to correct the deviation of the sensor so as to improve the accuracy and reliability of the data. According to the garage parking space safety risk monitoring method, static and dynamic risk data are processed by integrating data models such as a support vector machine and a random forest, physical risks are calculated in combination with a physical model, empirical rules are supplemented by an expert system, results are fused through a collaborative reasoning mechanism, and weights are dynamically adjusted, so that the evaluation accuracy and adaptability are improved; potential risks are accurately identified, and reliable support is provided for safe operation of the garage.

The present invention relates to a production expert system-based iron ore blending optimization method. The method comprises: firstly, performing data enhancement on historical ore blending and beneficiation data by means of an improved generative adversarial network; then, establishing a case library, and by means of a case retrieval method, establishing a relational model between blended ore properties and ore concentrate grades on the basis of the historical ore blending and beneficiation data and expert experience; then, on the basis of material balance, establishing a blended ore property prediction mechanism model to realize blended ore property estimation after ore blending; and finally, by taking an optimal ore beneficiation concentrate grade and a maximum difficult-to-beneficiate ore use amount as optimization objectives, determining constraints on the basis of actual working conditions, obtaining a set of ore blending proportion non-inferior solutions by means of a multi-objective grey wolf algorithm, and deciding an optimal solution by means of a TOPSIS algorithm. The present invention has the following advantages: while ensuring the quality of ore beneficiation products, reasonable ore blending proportions are calculated, the use amount of difficult-to-beneficiate ores is increased, the sustainable mining capacity of mines is improved while ensuring the mining and beneficiation benefits, and the invention has universal applicability.

This disclosure describes an expert system that can be used to automatically understand the function of a binary. The expert system includes a large language model (LLM) to determine investigatory steps that are implemented by a suite of tools. One application is malware detection. The expert system uses the tools to gather data and manipulate the binary to gain greater understanding of its function. Data generated during the investigation can be stored and retrieved from a memory representation system. This involves the LLM designing an investigation plan based on both default choices and responses to the data gathered using the tools. The expert system can adjust the plan after each step. Translators use expert knowledge and understanding of tool functions to convert tool outputs into natural language prompts that can be meaningfully understood by the LLM and to convert natural language output by the LLM into calls to the tools.

The present invention discloses a modular CSTR fermentation simulation equipment and an adaptive control method, which relate to the field of biological fermentation and process control. The equipment is composed of modules such as a detachable fermentation unit, a gas monitoring unit, and a data acquisition unit, and integrates multi-parameter sensors and compensation algorithms to achieve closed-loop control. The control method adopts a two-layer cascade architecture: the lower-level controller stabilizes the pH by adjusting the feed rate through dynamic PID; the upper-level controller dynamically optimizes the pH set value based on the gas production deviation, and forms feedback control in combination with the extreme value search model. The expert system periodically analyzes the gas production deviation, detects the load limit through micro-disturbance, and dynamically adjusts the target parameters to achieve organic load optimization. The present invention integrates multi-parameter collaborative monitoring and intelligent control models, calculates key parameters such as OLR and HRT in real time, improves processing efficiency while maintaining system stability, breaks through the hysteresis defects of traditional manual control, and provides an adaptive solution for anaerobic fermentation processes.

A method and a system of configuring custom operator defined rules for managing radio resources of user equipment in a wireless communication system. The system includes a plurality of User Equipment (UE), a plurality of wireless base stations configured to host Centralized Unit (CU) and Distributed Unit (DU) software, a rules engine, a first Application Programming Interface (API) wherein the rules engine is programmed with operator defined rules via the first API, and a second API coupled with the CU and DU software and configured to invoke the rules engine upon an occurrence of one or more defined events, wherein one or more of the operator defined rules is invoked via API calls to execute operator defined rules that match one or more inputs provided by at least one of the CU and DU hosted by at least one of the plurality of base stations.

The invention discloses an equipment pressure testing method, equipment, a storage medium and a program product, and relates to the technical field of pressure testing, and the method comprises the steps: constructing pressure testing strategies of a first system and a second system in to-be-tested equipment, and executing corresponding pressure testing operation; collecting first test data of the first system and second test data of the second system, and extracting first feature data of the first test data and second feature data of the second test data; generating a target feature based on the first feature data and the second feature data; and generating a diagnosis result based on the target feature by using a diagnosis model constructed based on an expert system and a machine learning model, and generating a diagnosis report based on the target report template. According to the method, the pressure test data of each operating system can be collected, the features are extracted and fused, the model fusing the expert system and machine learning is used for correlation analysis and automatic diagnosis, automatic and accurate positioning of the performance bottleneck is realized, and the diagnosis comprehensiveness and accuracy are improved.

The invention provides a power plant equipment sound abnormity identification and health prediction method based on granular computing and an LSTM network, and the method comprises the steps: employing an array pickup and a high-dynamic-range microphone array in a complex and high-noise background environment of a power plant, and combining an anti-interference filtering and beam forming algorithm, thereby achieving the sound abnormity identification and health prediction of the power plant equipment. According to the method, directional, multi-channel and non-contact sound acquisition is carried out on key parts of equipment, acquired sound signals are preprocessed, converted into time domain, frequency domain and time-frequency domain representations and coded into multi-dimensional numerical vectors, and compared with a rule-based expert system, the method has higher generalization ability and real-time response ability; a large language model is introduced, so that the output is closer to an engineering language and is suitable for operation and maintenance personnel to understand and execute; a self-defined knowledge base or safety semantic filtering is supported, and closed-loop deployment in an industrial field is facilitated; the system can be in butt joint with an intelligent inspection system, and full-link linkage of voice recognition, health assessment and suggestion generation is achieved.

The invention discloses an open vocabulary target detection method and system based on mixed expert fine tuning. According to the invention, a Grounding DINO model is used as a pre-training model, and a hybrid expert (MoE) module is respectively introduced into a feedforward neural network module in an original decoder and a cross attention module in a feature enhancement module, so that different expert combinations can process tokens of different semantic information, thereby enhancing the open vocabulary detection capability of the network. According to the method, the hybrid expert system is introduced into the Grouping DINO model for the first time, and experimental results on a data set show that the open vocabulary detection capability of the model obtained through fine tuning is improved. Meanwhile, the method disclosed by the invention can also be used as a plug-in to be added into other Transform structures, and excessive reasoning overhead cannot be generated.

The invention discloses a routable distributed fingerprint embedding and migrating method and device based on hybrid experts, and belongs to the technical field of large language model intellectual property protection. The basic model is transformed based on the hybrid expert system architecture, the basic model is divided into a plurality of expert sub-modules and a gate control module, at least one expert sub-module is designated as a fingerprint expert module, and the gate control module is used for routing the trigger sample to the corresponding fingerprint expert module; the fingerprint expert module is finely adjusted based on the fingerprint data set, and model parameters embedded with fingerprint features are stored in an independent low-rank adapter; combining the fine-tuned fingerprint expert module and the low-rank adapter thereof with the expert sub-module of the corresponding level in the downstream model to realize fingerprint migration; and inputting the trigger sample into the downstream model after fingerprint migration for copyright verification and evaluation. According to the method, the model performance, the security and the verifiability are considered, and the adaptability of the fingerprint technology in an industrial large model deployment scene is improved.

The invention relates to an optimization method and device of a hybrid expert system, computer equipment and a readable storage medium. The method comprises the steps that an input sequence is input into a routing operator, a first output tensor and a second output tensor are obtained through output, the first output tensor represents the weight value of each input mark in the input sequence corresponding to each feed-forward network, when the feed-forward network is not activated, the weight value of the feed-forward network in the first output tensor is 0, and when the feed-forward network is not activated, the weight value of the feed-forward network in the second output tensor is 0; the second output tensor represents the total number of the activated feed-forward networks and the index of each activated feed-forward network; and inputting the input sequence, the first output tensor and the second output tensor into each feed-forward network, outputting a third output tensor by the activated feed-forward network, and performing accumulation and summation processing on the third output tensor to obtain an output result that the hybrid expert system is consistent with the data shape of the input sequence. By adopting the method, the reasoning efficiency of the hybrid expert system in a reasoning scene can be improved.

The invention provides a sepsis multi-task prediction method and system, and belongs to the technical field of disease prediction based on a mathematical model, and the method comprises the steps: obtaining a symptom data record of a to-be-detected patient; and processing the obtained symptom data records by using a pre-trained classification model to obtain a final task result. According to the method, a multivariable time sequence in medical health record data of a patient is utilized, time representation containing sepsis pathological characteristics is obtained through a time mode of modeling data based on a self-attention encoder, and the trend of the condition of the patient changing along with time is obtained; correlating different medical features according to features including patient condition change trends, so that each feature can perceive other features mutually, and mining global feature representation of the patient; and inputting the global feature representation into a task-aware hybrid expert system, and routing the unified global feature representation into different expert systems, thereby obtaining task-level representation and realizing multi-task prediction.