291results about How to "Rich diversity" patented technology

The invention provides a multi-modal large model processing method and device, a storage medium and a program product. The method comprises the steps of obtaining an original image set of a monitoring scene; identifying feature information of a target object in the original image set; according to the feature information, generating question and answer pairs corresponding to the original image set; and taking the original image set and the corresponding question and answer pair as training data, and performing fine tuning training on a pre-trained multi-modal large model to obtain a target multi-modal large model adaptive to the monitoring scene. According to the method and the device, the multi-modal large model can better adapt to the requirements of a specific monitoring scene, the performance and the application effect of the multi-modal large model in the monitoring scene are improved, and the intelligent level of a monitoring system can be improved.

The invention discloses a bearing fault diagnosis method based on a parallel multi-fusion deep learning network model. And obtaining a sample data set, dividing the sample data set into a training data set and a test data set, carrying out parallel multi-fusion deep learning network model training by using the training data set, and inputting the test data set into the trained parallel multi-fusion deep learning network model to carry out fault diagnosis. Wherein the parallel multi-fusion deep learning network model comprises three parallel sub-networks, namely a bidirectional gating circulation unit sub-network, a fast Fourier transform-convolutional neural network sub-network and a discrete wavelet transform-enhanced depth separable ion network, and a dynamic multi-head attention mechanism module. According to the method, the three parallel sub-networks and the dynamic multi-head attention mechanism module are combined to automatically extract and identify the fault features of the rolling bearing under multiple working conditions, and self-adaptive fault feature extraction under different domain signals and rolling bearing fault diagnosis under complex working conditions are realized.

The invention provides a method for generating infrared-visible light data in pairs, a diffusion model and a system. The method comprises the following steps: obtaining a plurality of pairs of infrared features and visible light features; obtaining a plurality of fusion features; obtaining a plurality of hidden space features; obtaining a plurality of pairs of infrared-visible light image data after reconstruction; obtaining features after noise addition is completed; obtaining a loss function of a de-noising sub-model of the diffusion model; obtaining a trained diffusion model; performing feature fusion on a visible light image of a to-be-detected target and a text instruction, inputting the fused image and text instruction into a de-noising sub-model in the trained diffusion model, performing reasoning de-noising on the de-noising sub-model in the trained diffusion model, and outputting a de-noised hidden space feature; and decoding the denoised hidden space features to obtain paired infrared-visible light image data corresponding to the target to be detected. According to the invention, paired infrared-visible light data can be generated at the same time, and the high consistency of the two information is maintained.

The invention relates to the technical field of artificial intelligence, and discloses an anti-fraud model training method based on deep learning, which comprises the following steps: step 1, preprocessing and constructing multi-modal data, carrying out abnormal value processing, missing value filling and feature standardization on original data, representing processed transaction data and geographic information through a feature extraction module, and carrying out feature extraction; the equipment information and the social relation graph are embedded through an embedding layer and a graph neural network; step 2, based on data enhancement of self-supervised learning, pre-training unlabeled data by designing a behavior prediction task and a mask filling task so as to initialize a main model by using generated model parameters, and generating a synthetic sample extension training data set; by designing a behavior prediction task and a mask filling task, automatic learning and enhancement of features are realized by using label-free data, the dependence on labeled data is reduced, the detection capability of the model on fraudulent behaviors is remarkably improved, and the diversity of training data is enriched.

The invention discloses an operation site safety management and control method and system based on a multi-modal large model, and the method comprises the steps: collecting site violation data, and carrying out the preprocessing, and obtaining multi-modal data; constructing a multi-modal large model; carrying out training and optimization on the multi-modal large model by utilizing the preprocessed multi-modal data; and using the optimized multi-modal large model to realize operation site safety management and control. According to the invention, a visible light image, a safety violation knowledge base and other modal data are fused, and a comprehensive safety monitoring and management platform is constructed by using a large model and a big data processing technology, so that real-time monitoring, early warning and intelligent response of potential safety hazards of a complex operation site are realized, and the safety management level of the operation site is improved.

The invention relates to the technical field of lithium batteries, and discloses a lithium battery life decline quantitative prediction and health state intelligent evaluation method, which comprises the following steps: S1, multi-source data acquisition: arranging strain gauge sensors at battery electrode plates and a shell to acquire stress-strain data, integrating a gas sensor in a battery module to acquire gas data, and calculating the data of the stress-strain data; battery data and environment data are acquired by a high-precision synchronous acquisition system through nanosecond clock signals and are processed and stored, and the data quality is monitored at the same time; s2, data preprocessing: generating an extended data set through data transformation and combination and GAN-based data, and processing abnormal data by using a method based on statistics and machine learning; according to the invention, through multi-source data fusion acquisition, a stress-strain sensor and a plurality of gas sensors are added to monitor the internal mechanical change and gas component concentration of the battery, so that the problem that the accuracy and practicability of the model are greatly reduced due to a traditional method based on an electrochemical model is improved.

The invention relates to the field of medical model training, and particularly discloses a medical treatment guide model training method and system, a terminal and a medium, and the method comprises the steps: carrying out the incremental pre-training of a selected large language model through priori knowledge, and obtaining a basic large language model; acquiring a public real doctor-patient dialogue data set; constructing a doctor-patient dialogue template, and generating and constructing a doctor-patient dialogue data set by using a third-party large language model based on the doctor-patient dialogue template; combining the real doctor-patient dialogue data set and the constructed doctor-patient dialogue data set into a data set, and recording the data set as an initial doctor-patient dialogue data set; performing duplicate removal preprocessing on the initial doctor-patient dialogue data set to obtain a doctor-patient dialogue training data set; and using the doctor-patient dialogue training data set to carry out supervision fine tuning training on the basic large language model to obtain a doctor-seeing and hospital-guiding model. According to the invention, the joint training data set is formed based on the public data set and the automatic construction data set, so that the accuracy of the medical treatment guide model is improved.

The invention discloses a routing path determination method and device, a medium and a product, and relates to the technical field of communication. On the basis of a triangular prism network structure, when a source routing node and a target routing node are located on the same layer, the source routing node and the target routing node are determined by adopting a triangular prism side edge routing rule, a triangular prism surface routing rule and a relation of actual exclusive or results of the two routing nodes. According to the triangular prism side edge routing rule, side edge routing is preferentially carried out under the condition that the two routing nodes are not located on the same triangular prism face, the routing path is made to be short, and finally the two routing nodes move to the same triangular prism face. The routing rule of the triangular prism surface is based on the routing rule of routing along the triangular prism surface under the condition of the same surface, the flexibility and diversity of the routing path are determined according to different rules, and deadlock is avoided. And when in different layers, the system needs to jump to the same layer for routing planning, so that the reliability of communication is ensured.

The invention discloses a medical text data synthesis method based on role driving, and relates to the technical field of text data synthesis and artificial intelligence, and the method comprises the following steps: S100, generating an original problem; s200, screening original problems; and S300, synthesizing the medical text data. According to role description and interest content generation, the adaptability to different medical role requirements is enhanced, semantic duplicate removal and correlation evaluation are realized by utilizing a text coding model and a vector database, semantic duplicate content is removed, correlation is accurately evaluated by utilizing a large language model, the data quality and diversity are improved, and the method is suitable for popularization and application. Through generation of variant questions and answers, diversity and practicability of medical text data are enriched. According to the method, the personalization and pertinence of the medical text data are improved, the quality and duplicate removal efficiency of the medical text data are enhanced, the problem types are enriched, the generated content is optimized, and various information requirements of the medical text data are comprehensively covered.

The invention discloses a data processing system suitable for accounting financial management, which relates to the technical field of financial data processing and is characterized in that accounting financial management demand tasks are acquired through communication connection with a user side, and real-time data streams are screened according to association matching information between the real-time data streams and the accounting financial management demand tasks, so that the real-time data streams are acquired; obtaining task first-level data; searching historical calling data of the accounting financial management demand task, and screening the historical calling data of the accounting financial management demand task according to timeliness and usability standards of the data to obtain task secondary data; calling task third-level data based on the degree of correlation with the task second-level data; and integrating the task first-level data, the task second-level data and the task third-level data to obtain a task demand data set, and then feeding back the task demand data set to the user side. The optimization of the data screening process is realized, the real-time data processing capability and the data analysis precision are improved, and the historical calling data are efficiently utilized.

The invention discloses a data conversion method and device based on machine learning, a medium and equipment, and the method comprises the steps: obtaining original data, and carrying out the preprocessing of the original data; performing feature selection on the preprocessed data based on a feature selection algorithm; inputting the data after feature selection into a corresponding trained rule generation model based on a data type to generate a data conversion rule; wherein the rule generation model is constructed and trained according to different data types; acquiring a real-time data stream, optimizing the rule based on the trained rule optimization model, and / or adaptively adjusting the rule according to the real-time monitoring data; and processing the data through the adjusted rule, and performing anomaly detection and correction on the data after rule processing. According to the method, the automatic rule is generated through real-time data pattern recognition, rule optimization and comprehensive data detection and correction are carried out in a self-adaptive mode, and the efficiency and accuracy of data processing are remarkably improved.

The invention discloses an AGV obstacle avoidance path planning method in an unmanned warehousing system, and the method comprises the steps: employing nested Logistic mapping and Skew Tent mapping to generate a chaotic initialization population to replace a population generation mode of a conventional differential evolution algorithm, carrying out the Levy flight operation after the selection operation, and carrying out the global path search on an environment map, thereby achieving the optimal path planning. And obtaining an AGV global feasible path avoiding obstacles in the environment map, and when the AGV has an unknown obstacle on the feasible path, re-planning the local path by adopting a dynamic window method so as to enable the AGV to realize an obstacle avoidance process in an interval where the unknown obstacle is located. The improved differential evolution algorithm enhances the global search capability through chaos initialization and a Levy flight strategy, and the AGV can quickly respond to an unknown obstacle through a dynamic window rule. The adaptability and the accuracy of path planning are improved, and the obstacle avoidance capability and the safety of the AGV in a complex environment are enhanced at the same time.

The invention provides an AI-based three-dimensional medical image intelligent labeling method and system. The accuracy and efficiency of medical image labeling are remarkably improved. By combining an image classifier of preliminary knowledge learning with a basic lesion labeling network, knowledge information from different sources can be synthesized, and a more comprehensive and reliable lesion semantic knowledge point sequence is generated. Furthermore, through three types of training image generation strategies, complementarity among different networks is effectively utilized, the recognition capability of common lesion areas is enhanced, and lesion areas which are respectively and uniquely found are explored, so that diversity and representativeness of training data are enriched. Finally, deep knowledge learning is carried out on the basic lesion labeling network based on the diversified training data, the recognition precision and generalization ability of the generated target lesion labeling network are remarkably improved, the workload of manual labeling of doctors is relieved, and the medical image analysis process is accelerated.

The invention belongs to the technical field of image processing, and discloses a loess landslide detection method and system optimized by a multi-population grey wolf optimization algorithm based on grade differentiation learning. Performing data enhancement on the landslide data set by using a data enhancement generator; designing a grey wolf optimization algorithm model based on multi-level differential learning through the segmentation network model, and optimizing hyper-parameters in the segmentation network model by using an improved algorithm to realize adaptive adjustment and cooperative calculation of three hyper-parameters, namely a learning rate, an iteration number and a batch processing number, in the segmentation network; and finally carrying out loess landslide detection. According to the method, algorithm improvement is carried out, so that the balance capability of global exploration and local development is enhanced, and later diversity loss and local optimum are avoided. By combining the excellent semantic segmentation capability of VIG and the improved grey wolf optimization algorithm, the method can effectively overcome the feature extraction difficulty caused by factors such as geological structure and vegetation coverage, and significantly improves the accuracy of loess landslide recognition.

The invention discloses a drug molecule virtual screening method based on combination mode analysis, and belongs to the field of artificial intelligence drug discovery. Using PLIP to calculate binding modes between the to-be-screened target and the small molecules, performing coding, using a DBSCAN algorithm to cluster the binding modes, and identifying compound populations of similar binding modes; in the screening stage, two key indexes are introduced: an energy index is formed according to a hydrogen bond compensation number and an AutoDock Vina score, the correlation between related residues and key site alpha carbon atoms in a binding mode is used as an activity index for evaluating the binding stability and characteristic activity between a target and a small molecule, and multi-target frontier analysis is performed on candidate molecules; according to the method, the efficiency and precision of virtual screening are remarkably improved, detailed analysis on intermolecular interaction is enhanced, the method is suitable for small-scale and large-scale molecular screening, the drug research and development cost is reduced, the drug research and development time is shortened, the diversity of virtual screening results is enriched, and potential candidate molecules are provided for subsequent experiments.

The invention discloses a non-contact real-time monitoring method for human respiration in a Wi-Fi (Wireless Fidelity) complex interference environment. And analyzing the CSI interference intensity and the CSI active ratio, constructing a Wi-Fi interference feature mapping matrix, and calculating each channel interference index by using the matrix to realize interference discrimination. Then, through a dynamic subcarrier selection algorithm CSI-DSSA based on an interference index, a subcarrier combination with the weakest cross correlation in interference data is selected for interference processing, and time domain feature information of multiple data streams in data which is not interfered in a multi-link data fusion method aggregation mode is analyzed. And finally, extracting a time domain characteristic value and constructing an SVM multi-activity classification model in a Wi-Fi interference environment to obtain a tumble activity identification result. According to the method, the human respiratory activity recognition accuracy in the Wi-Fi interference environment can be effectively improved.

The invention discloses a feature extraction method based on a multi-view interactive neural network, and the method comprises the steps: constructing a multi-scale convolutional neural network based on multiple modules, achieving the extraction of multi-scale feature information through a parallel multi-core, and designing the cross-layer connection of a plurality of network modules to achieve the feature interaction among the multiple modules. Information sharing is completed; constructing an interactive Transform network based on local and global features, applying a lightweight residual block in the network to realize local association of the global features, and applying a lightweight convolutional layer to replace a traditional convolutional block to reduce time consumption cost; features extracted by a multi-module multi-scale convolutional neural network and a feature interactive Transform network are fused, and the method is applied to multi-spectral target recognition. According to the method, global and local information is interactively interpreted in the network optimization process, so that feature information is richer, and finally, target recognition performance with relatively high precision is obtained.

The invention discloses a pedestrian re-identification pre-training method based on cross-video identity association, and belongs to the field of pedestrian re-identification, and the method comprises the steps: employing single tracking frame denoising to remove the noise in each single tracking frame in a cross-video pedestrian image set, employing single video denoising to remove the noise in each single video in the cross-video pedestrian image set, and then employing single tracking frame denoising to remove the noise in each single video in the cross-video pedestrian image set; removing noise in a plurality of videos in the cross-video pedestrian image set by adopting cross-video denoising to obtain the denoised identity tag of each pedestrian image in the cross-video pedestrian image set; pre-training the self-distillation learning network architecture by using each pedestrian image and the corresponding denoised identity tag to meet the consistency probability distribution constraint as a target; wherein the self-distillation learning network architecture comprises a teacher network and a student network, and the teacher network and the student network update network parameters in an index moving average mode. According to the method, the pedestrian re-identification pre-training model with higher representation capability is obtained.

The invention discloses a traffic police dynamic gesture recognition method and system based on vehicle-mounted federal learning, and the method enables a key feature extraction model and a preset global model to be combined as a traffic police gesture recognition model shared by vehicle-mounted terminals of a whole network, and facilitates the improvement of the gesture recognition precision of the traffic police. Different vehicle-mounted terminals are organized to train the models based on the collected traffic police gesture image data through vehicle-mounted federal learning, and then each vehicle-mounted terminal only shares model update data instead of original training data, so that the data privacy of vehicle users is protected, the communication cost for uploading mass data to a central server is saved, and the user experience is improved. Meanwhile, real-time traffic police gesture image data dispersed in different vehicle-mounted devices are fully utilized in vehicle-mounted federation learning, training data from vehicle-mounted images have different observation angles, different environment conditions and the like, the diversity of model training data is enriched, and the training efficiency is improved. And thus, the adaptive capacity of the trained traffic police gesture recognition model in a variable traffic environment is enhanced.

The invention relates to the technical field of adversarial learning, in particular to an adversarial sample generation device and method based on space-frequency fusion, and the method comprises the steps: collecting image data, and building a training data set for adversarial sample generation, including building a target class image training set and a to-be-generated disturbance image training set; performing spatial domain block data enhancement and frequency domain data enhancement on an image in the to-be-generated disturbance image training set to obtain a data enhanced image; processing the data enhancement image by adopting a generator to obtain a disturbance image; based on the images in the target class image training set, a discriminator is adopted to carry out space domain discrimination and frequency domain discrimination on the disturbance images, and space-frequency domain loss is obtained; taking minimization of the space-frequency domain loss as an optimization requirement, and updating network parameters of the generator; obtaining a trained generator until the space-frequency domain loss is smaller than a preset value, and generating an adversarial sample by adopting the trained generator; according to the method, the adversarial sample with better mobility can be generated.

The invention relates to the technical field of plasmon nano photonics, and discloses a preparation method of a novel gradient metasurface, which comprises the following steps: pretreating an aluminum sheet; prefabricating a pattern on the aluminum sheet and performing anodic oxidation; the preparation method comprises the following steps: preparing an anodized aluminum oxide thin film, fixing the anodized aluminum oxide thin film on a glass sheet, placing the glass sheet fixed with the anodized aluminum oxide thin film in chambering liquid, and taking out the glass sheet to obtain the novel gradient metasurface. By adopting electrochemical and chemical corrosion methods, for gradient metasurfaces with different areas, different sizes and different directions, the diversity of adjustable parameters of the nano array can be enriched, the time is shortened, the cost is reduced, the manufacturing cost and time are reduced to a great extent, and the production efficiency is remarkably improved.

The invention provides a model training method and device, a recommendation method and device and related equipment, and the model training method comprises the steps: determining a first training sample, the first training sample comprises first user attribute data of a first sample user, first behavior data of the first sample user on a first sample object and first object attribute data of the first sample object; determining a first behavior label and a user click rate of the first sample object by the first sample user according to the first behavior data, and determining an object click rate of the first sample object according to the first object attribute data; predicting a first behavior probability of the first sample user to the first sample object based on the first user attribute data, the first behavior data and the first object attribute data; determining a first loss value based on the first behavior tag, the first behavior probability, the user click rate and the object click rate; and training a first model according to the first loss value. The method can effectively improve the prediction accuracy of the first model.

The invention discloses a rusted high-strength bolt connection pretightening force intelligent self-optimization prediction method and system, and belongs to the technical field of bolt pretightening force detection.The method specifically comprises the steps that a finite element database of bolt connection pretightening force loss is established, and a machine learning prediction model based on a neural network is trained; when the recognition accuracy of the machine learning prediction model reaches a preset value, pre-tightening force loss of the target bolt is monitored in real time, and real-time geometric feature information of connection of the target bolt is obtained; when the similarity between the real-time geometric feature information of the target bolt connection and a finite element model in the finite element database is greater than or equal to a preset value, calling the pre-tightening force loss of the existing finite element database; when the similarity is lower than a preset value, the trained machine learning prediction model is utilized to output actual bolt connection pre-tightening force loss in real time; the bolt looseness detection device realizes automatic and marker-free bolt looseness detection, is high in bolt image acquisition angle and distance tolerance, and can meet bolt looseness detection at different positions.

The invention relates to the technical field of railway bridge crack detection and positioning, in particular to a steel plate crack positioning method based on a convolutional neural network and a particle swarm optimization algorithm, which comprises the following steps: drawing grid lines in a corresponding region of a steel plate, detecting acoustic emission signals of specific coordinate positions in the corresponding region of the steel plate based on a convolutional neural network model, obtaining a steel plate crack first positioning result; wherein the convolutional neural network model adopts a global mean pooling layer to execute average pooling operation on each channel in the two-dimensional feature map after convolution processing, and outputs a one-dimensional feature vector; or introducing an adjustable learning factor and an inertia weight into a particle swarm optimization algorithm, and carrying out traversal search on the acoustic emission signals of the corresponding areas of the steel plate in combination with the crossing thought of a genetic algorithm to obtain a second positioning result of the steel plate crack; and taking the first positioning result or the second positioning result as a final positioning result of the steel plate crack. According to the invention, automatic detection and accurate positioning of bridge fatigue cracks can be realized.

The invention discloses an automobile part picture recognition method, which comprises the following steps: acquiring an automobile part picture, and performing data cleaning processing on the automobile part picture to obtain a plurality of picture subsets; performing data annotation processing on the plurality of picture subsets to obtain picture training data; inputting the picture training data into a model constructed based on a ResNet network for training to obtain an accessory picture recognition model; performing global feature extraction on each picture in the picture training data through a preset global feature extraction model to obtain a global feature vector set; obtaining a to-be-recognized picture, and predicting the to-be-recognized picture through the accessory picture recognition model to obtain a recognition result; and when the identification result is judged to be a commodity picture, taking the identified accessory commodity picture as a return result, otherwise, by calculating the Euclidean distance between the global feature vector corresponding to the picture to be identified and each vector in the global feature vector set, determining an image retrieval result according to the Euclidean distance, and returning the image retrieval result to the user.

The invention provides a distributed flexible workshop scheduling method and system based on a multi-target grey wolf algorithm, and belongs to the technical field of workshop scheduling. The method comprises the following steps: initializing algorithm parameters, and defining a coding and decoding mechanism by adopting three-layer coding including a process sequence, machine selection and factory selection; initializing the initial wolf pack by adopting a hybrid strategy based on three-layer coding; dividing non-dominated levels of the initial wolf pack through non-dominated sorting and sorting population crowding degree; determining the position of a wolf pack leader based on the number of non-dominated hierarchies and crowding degree sorting; performing crossover variation and wolf pack combination on the individual positions of the wolf pack, and updating the combined wolf pack to obtain a non-dominated elite solution set; setting a destruction operator and a repair operator, dynamically adjusting the selection probability of the destruction operator and the repair operator, introducing a simulated annealing algorithm to expand a search solution space, and optimizing an elite solution set to obtain a final solution. The scheduling efficiency of the distributed flexible job shop is improved, local optimum is avoided, and the production benefit and quality are guaranteed.

The invention belongs to the technical field of seed treatment, and particularly relates to a method for promoting germination of coriander seeds. The method for promoting the germination of the awngrass seeds comprises the following steps: before the germination of the awngrass seeds, pretreating the awngrass seeds by using a seed soaking agent; the seed soaking agent is a solution containing a plant growth regulator; the concentration of the plant growth regulator in the seed soaking agent is 5mg / L to 160mg / L. According to the method, seeds are soaked in a seed soaking agent, so that dormancy of the miscanthus coriander seeds is broken, and germination of the miscanthus coriander seeds is promoted. According to the method provided by the invention, the germination speed, the germination rate, the germination potential and the germination index of the miscanthus coriander seeds are remarkably improved. In addition, as the seed soaking agent can be used for repeatedly soaking the seeds, the method also has the characteristics of simplicity in operation, time and labor saving and low cost. Therefore, the method disclosed by the invention has important application and popularization values for the technology of breaking dormancy of the miscanthus coriander seeds and promoting germination of the miscanthus coriander seeds.

The invention provides an infrared image enhancement method based on pseudo noise and a convolutional neural network, and the method comprises the steps: collecting high-definition visible light image data, carrying out the cleaning, graying and fuzzification of the collected image data, adding pseudo noise, and building an initial data set; performing data enhancement on the obtained initial data set, and establishing a sample data set comprising a training set, a verification set and a test set; constructing an infrared image enhancement network based on a convolutional neural network, and training and optimizing the network by using the sample data set to obtain an optimal infrared image enhancement model; dividing an actually collected infrared image into three channels, inputting the three channels into the infrared image enhancement model for enhancement, and merging the results of the three channels to obtain a final infrared enhanced image. The method can effectively remove the noise of the infrared image, enhances the target features in the infrared image, and enables the high-quality infrared image to be obtained under a complex environment condition.

The invention discloses a P-glycoprotein inhibitor and substrate prediction method based on multi-modal contrast learning. The method comprises the following steps: constructing a data set of a P-glycoprotein inhibitor and a substrate; constructing a P-glycoprotein inhibitor and substrate prediction model by using a deep learning method; inputting the comparison loss and the classification loss into the constructed P-glycoprotein inhibitor and substrate prediction model, combining the comparison loss and the classification loss to obtain a total loss function, and training and optimizing parameters of the P-glycoprotein inhibitor and substrate prediction model through the total loss function; and inputting data of a compound to be predicted into the optimized P-glycoprotein inhibitor and substrate prediction model, outputting a final classification result, and judging whether the compound is the P-glycoprotein inhibitor and the substrate or not. According to the method, the prediction precision and generalization performance of the model are improved, the interpretability of the model is realized, and a reliable scientific basis is provided for molecular design in drug development.

The invention discloses a video editing system for artificial intelligence teaching, which comprises an editing source material determination module, a source material content learning determination module, a video editing mode determination module, a video editing execution module, an editing result evaluation module and a user interaction module, and relates to the technical field of artificial intelligence. According to the video editing system for artificial intelligence teaching, through integration of intelligent screening, classification, deep learning and automatic editing technologies, operations of preliminary screening, content analysis, editing mode determination, special effect addition and the like of a large number of video materials can be automatically completed. Therefore, the workload of manual editing is greatly reduced, and the accuracy and efficiency of video editing are remarkably improved. Meanwhile, the system supports touch operation and voice control, the editing process is further simplified, and the teacher can make high-quality teaching videos more quickly and conveniently.