17results about How to "Reduce sparsity" patented technology

The invention relates to a shopping center shop recommendation method and device, equipment and a medium, and the method comprises the steps: predicting interested shops of a target old user in a current time period through employing a long and short-term memory network model trained by user behavior time sequence data, adopting a preset collaborative filtering algorithm to calculate and determine a second similarity between the target old user and other users according to the interested shops of the target old user in the current time period, so as to determine neighbor users of the target old user; predicting an estimated score value of a target old user for an unscored shop according to the score value of the neighbor user for each shop, and calculating and determining a second recommendation degree corresponding to each shop according to the estimated score value, the actual distance between the target old user and the shop and the comprehensive recommendation value, and displaying each shop to a user interface of the target old user according to the second recommendation degree from high to low so as to complete shop recommendation of the shopping center. According to the invention, the traffic and user stickiness of a shopping center can be improved.

This invention provides a drug interaction prediction method based on a bidirectional cross-view attention network, belonging to the field of drug interaction prediction technology. The method first constructs a Morgan fingerprint similarity view of the drug, an original DDI view, and a multi-scale diffusion view based on personalized PageRank. Then, a graph convolutional network with shared weights is used to co-encode the multiple views, generating a unified drug embedding representation. Finally, a bidirectional cross-view attention mechanism is used to achieve fine-grained interaction and alignment between the structural and attribute views, and interaction prediction is completed via a multilayer perceptron. This invention effectively alleviates the sparsity problem of the DDI network through multi-scale topology enhancement and achieves complementary enhancement between views using bidirectional cross-view attention, significantly improving the accuracy and generalization ability of drug interaction prediction. It can provide reliable technical support for drug development screening and clinical combined drug safety assessment.

According to the 3D target detection method based on fusion of the camera and the enhanced 4D millimeter-wave radar, a simple and effective point cloud enhancement algorithm is designed, and the sparsity problem of point clouds is relieved by adding effective four-dimensional radar point clouds; a point cloud feature fusion enhancement module is developed, target detail features are enhanced by constructing sparse voxels of different scales, and the system robustness is improved.

The present invention proposes a mobile user trajectory prediction method and system, which relates to the field of computer information communication and service computing technology, including: obtaining historical spatiotemporal trajectory data of mobile users, introducing social relationship data, and constructing a heterogeneous graph; through a feature extraction model based on contrastive learning, data enhancement is performed on the heterogeneous graph from a structure layer and a feature layer respectively, and the dependency relationship and interest preference in the spatiotemporal trajectory data are captured to extract trajectory features; based on the trajectory features, points of interest are predicted; the structure layer uses a singular value decomposition method to extract structural information in the heterogeneous graph and reconstruct the heterogeneous graph; the feature layer adds noise to the feature space during the contrastive learning process of the original heterogeneous graph and the reconstructed heterogeneous graph through a lightweight graph convolutional neural network; the present invention accurately extracts trajectory features and alleviates the sparsity problem of spatiotemporal data by introducing social relationship data and a contrastive learning mechanism for data enhancement from two levels.

A friend recommendation method based on cross-view comparative learning in a location social network belongs to the technical field of recommendation, and comprises the following steps: acquiring social platform data of a user, and constructing the location social network according to the social platform data; generating a distance constraint meta-path instance and a time constraint meta-path instance; performing cross-view comparison on the fine-grained view; performing cross-view comparison on the coarse-grained view; calculating final comparison loss, and training a comparison learning model according to the final comparison loss to optimize user node embedding; and calculating the similarity of the users, and recommending friends according to the similarity. According to the method, the friend recommendation data density is improved by using time and space constraints, and the consistency and complementarity of heterogeneous information are captured at the same time so as to improve the recommendation accuracy.

The invention relates to the technical field of image processing, and discloses an operation scene real-time dynamic reconstruction method and system based on three-dimensional Gaussian sputtering. The method comprises the following steps: extracting spatial-temporal characteristics of a soft tissue deformation Gaussian subset through a HexPlane spatial-temporal encoder; fusing the decoded feature vectors to generate a soft tissue deformation parameter set; a deformation field Jacobian matrix of Gaussian elements in the soft tissue deformation Gaussian subset is calculated, a local volume strain rate is obtained through determinant operation, when the strain rate exceeds a preset threshold value, self-adaptive densification operation is executed, and new Gaussian elements inherit parent Gaussian element deformation parameters and are updated to a corresponding set; and combining the updated parameters to carry out probabilistic boundary allocation and transmissivity perception rendering to obtain a reconstruction result. The accuracy, the physical authenticity and the stability of real-time dynamic reconstruction of the operation scene are improved.

This invention belongs to the field of technology for technology achievement promotion, and particularly relates to a method for mining potential demand for technology achievements based on hypergraph reconstruction technology, including: S1, constructing a demander-technology achievement interaction graph; S2, constructing a hypergraph association matrix of demanders and a hypergraph association matrix of technology achievements; S3, using a graph neural network to capture the direct dependencies between demanders and technology achievements; using a hypergraph neural network to capture the global implicit dependencies between demanders and technology achievements; S4, combining the direct dependency encoding of demanders at each layer with the global implicit dependency encoding to obtain enhanced embeddings of demanders and technology achievements at each layer; S5, based on and calculating the demander S... i Select each scientific and technological achievement C j The probability prediction value is calculated, and the top k scientific and technological achievements are used as the recommendation results. This method can effectively address the problem of data sparsity.

The invention belongs to the technical field of internet application, and particularly relates to a social platform guide topic key node tracing method, which comprises the following steps: acquiring user data; preprocessing the user data; extracting attribute features of the preprocessed user data; processing the user data by adopting a hybrid random walk strategy to obtain a user hidden relationship; performing sentiment analysis on topic information in the user data to obtain user sentiment features; using a URE2vec algorithm to carry out vectorization representation on the user attribute features, the user hiding relationship and the user emotion features, and carrying out dimensionality reduction on a vectorization representation result to obtain hidden information; inputting the hidden information into a key node traceability model to obtain a topic propagation key node and a topic guide intention; according to the method, a hybrid random walk strategy is adopted, implicit friends with similar preferences are found out for the user through the hybrid random walk strategy by utilizing an explicit relationship and an implicit relationship, and data sparsity is relieved.

The application relates to a federated learning assisted edge caching method based on an AE-DDPM model, and comprises the following steps: an edge computing network system model is established, including a base station, a remote cloud server and users; an AE-DDPM model based on federated learning is used for training; global prediction content popularity is obtained, and the most popular N contents are cached according to the cache capacity of the base station; the application extracts a user data potential feature vector through an AE model, reduces data dimension and sparseness, and then learns data distribution through a DDPM model to generate high-quality content popularity prediction; the application deploys a cache unit on an edge node, enables users to quickly obtain the pre-cached contents on the node, effectively improves the cache hit rate, reduces the time delay of the users in obtaining the contents, significantly improves communication efficiency, and simultaneously reduces the risk of user privacy leakage.

The invention relates to the technical field of artificial intelligence, can be applied to the medical field and the financial science and technology field, and discloses a graph migration-based recommendation method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining original business data, constructing a heterogeneous graph based on the original business data, and obtaining auxiliary features; mapping the initial features and the auxiliary features of the nodes in the heterogeneous graph to obtain graph node embedding and auxiliary feature embedding; semantic information in auxiliary feature embedding is injected into graph node embedding to obtain migration enhancement graph embedding; embedding the migration enhanced graph to carry out graph volume accumulation to obtain enhanced graph features; embedding and fusing the enhanced image features and the auxiliary features to obtain target fusion features; and predicting preferences of the user on the articles based on the target fusion features to obtain preference information, and recommending the articles to the user based on the preference information. According to the method, the problem of data sparsity is effectively relieved, and the accuracy and robustness of a recommendation system in a cold start scene are improved.

The invention provides a point cloud data sparse convolution training method and system. The method comprises the steps of obtaining to-be-trained point cloud data; performing preprocessing operation on the point cloud data to obtain a target convolutional layer; disassembling the target convolutional layer into a convolutional layer group; the convolutional layer group comprises a plurality of convolutional layers of which the size is smaller than that of the target convolutional layer; extracting features of the convolutional layer group; performing normalization processing on the features to obtain numerical values and threshold values corresponding to the features; screening the numerical values by using the threshold value to obtain a target numerical value; and obtaining a target feature based on the target numerical value, so as to solve the problems that when feature extraction is performed on the point cloud data at present, sparsity of the point cloud data disappears, so that the calculation amount of the point cloud data is very huge, the training duration and size of a model are greatly increased, and the risk of overfitting is improved.

The invention discloses a multi-hop reasoning method and system based on type perception aggregation. Semantic information is achieved by constructing a type perception aggregation mechanism, expression extraction of entities and relation types is improved, and semantic content of an original knowledge graph is enriched. And by adopting a novel fusion mechanism, the representations of the entities and the relationship types are fused together more effectively, and the effectiveness of feature learning is further improved. In order to emphasize relation characterization highly related to the process, an enhanced relation attention mechanism is introduced, the relation characterization is amplified through Heo, Hso and Hro together, relation information can be aggregated in the node interaction process, therefore, updated node embedding has rich information, semantic information loss in the reasoning process is avoided, and the reliability of the system is improved. And the reliability of the reasoning result is ensured.

The invention belongs to the technical field of recommendation methods, and particularly relates to a graph contrast learning recommendation method based on two denoising modules. The method comprises the following steps: S1, based on a user-commodity bipartite graph, respectively using a first denoising module and a second denoising module to generate a new first view and a new second view; s2, respectively training the original view, the first view and the second view by using a LightGCN model; s3, using a BPR loss function to optimize a main recommendation task and an InfoNCE loss function to optimize a graph comparison learning task, and performing joint training to optimize overall loss; and S4, constructing a prediction layer, and obtaining a prediction result through the learned inner product of the final embedded representation of the user and the commodity. According to the graph contrast learning recommendation method based on the two denoising modules, the noise problem and the data sparsity problem can be effectively relieved, and the model recommendation accuracy and the model robustness are improved.

The present invention relates to the technical field of electric power systems, and discloses a GIS fault diagnosis method and system based on acoustic monitoring, including: obtaining initial sound data of each monitoring point in the GIS; performing a first fault location according to the initial sound data to obtain the area where the fault source in the GIS is located; and extracting the positioning sound data; performing a second fault location according to the positioning sound data to obtain the location information of multiple fault sources in the GIS; and performing voiceprint feature analysis according to the positioning sound data to obtain the fault types of multiple fault sources in the GIS; the fault types include partial discharge faults and mechanical faults; and performing GIS fault diagnosis according to the location information of the multiple fault sources and the fault types of the multiple fault sources. The GIS fault diagnosis method and system based on acoustic monitoring of the present invention significantly improves the accuracy of determining the location of the fault source, and simultaneously realizes the timeliness diagnosis of multiple fault sources.

The present invention provides a method, system, storage medium and electronic device for recommending time-aware hybrid cloud services based on entropy, and relates to the technical field of cloud service recommendation. The present invention includes: time-aware co-evaluation service weight calculation based on entropy; user similarity and trust calculation; QoS indicator score prediction and service recommendation. The cloud service recommendation method can effectively solve the problems of low service QoS prediction accuracy, poor recommendation effect, and data sparsity. Experiments have proved that introducing entropy-based service weights and user trust into traditional recommendation algorithms is feasible and effective, and has certain practical significance.

The application discloses a power battery field reliability evaluation method and device, equipment and medium, and relates to the field of power battery reliability evaluation. The method comprises the following steps: obtaining panel data of the power battery from the dimensions of working environment and working load; obtaining observation values of target influence factors determined based on influence weights; classifying geographical regions in a target area based on the panel data by using an improved deep clustering algorithm; constructing a relationship mapping model representing the relationship between field reliability influence factors and field reliability; determining fitting data according to the observation values of the target influence factors and the regional classification results, and fitting the parameters in the relationship mapping model based on the fitting data to obtain a field reliability evaluation model; and using the field reliability evaluation model to evaluate the field reliability of unknown area power batteries to obtain field failure data of the unknown area power batteries. The application can realize accurate evaluation of the field reliability of power batteries in different service conditions.

The invention relates to a privacy protection recommendation method and system based on matrix decomposition, and belongs to the technical field of privacy protection recommendation. Comprises: obtaining a historical score matrix of a user; performing matrix decomposition on the historical score matrix to obtain a shared score matrix; sending the shared scoring matrix and the scored item set to a recommendation server, and generating public parameters, a decomposition matrix and a decomposition matrix shared value; generating an inner product triple and a multiplication triple by using a triple server; calculating an error between the real score and the predicted score by using the recommendation server, and generating an error sharing matrix; the error sharing matrixes are optimized until the error sum of squares of all the error sharing matrixes is smaller than a threshold value; calculating a partial derivative of the loss function to the decomposition matrix to obtain a gradient sharing matrix; updating the decomposition matrix to obtain a user decomposition shared matrix; and generating a privacy protection recommendation result of the user by using the user decomposition sharing matrix. According to the invention, powerful guarantee is provided in the aspect of protecting user privacy.