21results about How to "Solve the sparsity problem" patented technology

The invention belongs to the field of medical ICD automatic coding, and particularly relates to a medical ICD coding classification method based on a graph neural network, and the method comprises the steps: obtaining a data set, and carrying out the preprocessing of the data set, and obtaining a text vector matrix; constructing a graph structure according to the text vector matrix, wherein the graph structure comprises a node feature matrix and an adjacent matrix; inputting the graph structure into a graph neural network model, outputting a prediction ICD coding result, and training the model by adopting a multi-label loss function based on cross entropy; inputting to-be-coded medical record data into the trained graph neural network model to obtain a predicted ICD code; according to the method, the semantic similarity between the medical terms, the level distance between the ICD codes and the co-occurrence frequency relation are included into a graph structure, the complex relation between words and labels can be comprehensively captured, the potential relevance between the ICD codes is better utilized, and the context understanding ability and accuracy of a classification model are enhanced.

The invention relates to a Chinese-blind conversion method of Transform-GRU based on consecutive character vector fusion, and belongs to the field of natural language processing and generation. According to the method, firstly, a word embedding technology for optimizing word segmentation continuous writing rules is realized through a multi-layer attention linear network, it is guaranteed that a model learns word vectors containing rich word segmentation rules, and compared with a general embedding technology, the method has higher pertinence for Chinese-blind translation; a Markov word segmentation model based on Gaussian normalization is designed, accurate word segmentation is achieved, and continuous word vectors are fused; according to the encoder design, high-level semantic information extraction is realized by an improved encoder formed by stacking a plurality of unidirectional extension layers and local attention layers, so that the problem of high complexity of a Transform network is improved, and targeted optimization is realized on Chinese and braille alphabets with consistent word orders; and the decoder designs a mask grouping self-attention mechanism and a GRU layer fused by bidirectional mutual attention gating, so that the capability of capturing rich semantics of coding information and the reasoning capability of the network can be improved.

The disclosed embodiments use a diffusion-based generative model for radar point cloud super-resolution. The disclosed embodiments use diffusion modeled arithmetical operations to generate higher resolution radar point cloud data from lower resolution radar point cloud data.

The invention provides a recommendation method and device based on hypergraph and multi-behavior contrast learning, equipment and a medium, and the method comprises the steps: obtaining interaction data of a user and an article, carrying out the preprocessing, and obtaining user behavior data and article attribute data; constructing a hypergraph for reflecting multi-behavior interaction and semantic information between the user and the article; obtaining feature representations of the user and the article based on the hypergraph, and enhancing the feature representations of the user and the article through multi-behavior contrast learning; and performing feature fusion based on the enhanced features, and performing target behavior prediction and recommendation. According to the method, information is enhanced through the hypergraph, and the relevance between the articles and the relevance between the users are captured. According to the method, more abundant and real feature representation is obtained based on feature comparison learning of behavior enhancement, and feature fusion is performed among different behaviors, so that the problems of data sparsity, behavior imbalance and the like are effectively solved, the recommendation accuracy and efficiency are improved, the user time is saved, and the user experience is improved.

The invention relates to a method for improving 2D laser radar point cloud data based on adaptive Fourier decomposition, and the method comprises the following steps: carrying out the interpolation processing of obtained original low-precision 2D laser radar point cloud data, so as to match the data density of high-precision 2D laser radar point cloud data; converting the interpolated low-precision 2D laser radar point cloud data into a linear curve, performing decomposition through adaptive Fourier decomposition, and extracting M components; the M components are divided into N groups, each group comprises K components, one component group is gradually removed from the high-frequency component, and the remaining components are combined to obtain second low-precision 2D laser radar point cloud data; calculating a root mean square error between the second low-precision 2D laser radar point cloud data in different elimination stages and the obtained original high-precision 2D laser radar point cloud data; comparing root-mean-square errors corresponding to different rejection stages, and selecting a component combination of the root-mean-square errors as an optimal component combination; and recombining the low-precision 2D laser radar point cloud data by using the optimal component combination.

The invention belongs to the technical field of commodity recommendation, and discloses an e-commerce commodity recommendation optimization management system, which comprises the steps of collecting e-commerce comprehensive data, and preprocessing the collected e-commerce comprehensive data to generate a comprehensive data set; weight values are distributed for the user behavior characteristics through a comprehensive data set; in combination with the weights of the user behavior characteristics, scores of the users for all the commodities are calculated, and a user-commodity interaction matrix is constructed according to the scores; the festivals and the user positions are comprehensively considered, the preference degree of the users for the commodities is calculated through the user-commodity interaction matrix, and the optimal commodity category recommendation list for attracting the users is generated, so that the problems of improper regional difference processing and difficult recommendation system performance optimization are solved, the commodity recommendation accuracy is improved, the user experience is enhanced, and the user experience is improved. And the user experience and the operation efficiency of the e-commerce platform are improved.

The invention discloses a single-effect lithium bromide absorption type water chilling unit design method and device, equipment and a medium, and the method comprises the steps: building a hydraulic and thermal coupling calculation model based on the thermodynamic characteristics of a single-effect lithium bromide absorption type water chilling unit; and obtaining input variables, inputting the input variables into the constructed hydraulic and thermal coupling calculation model, and outputting various operation parameters of the water chilling unit. According to the method, various operation parameters of the water chilling unit can be calculated based on external input variable coupling, the inaccuracy of a traditional design calculation method is avoided, and the problems of complex calculation and low precision of the single-effect lithium bromide absorption type water chilling unit under the multivariable coupling condition are solved. The invention can be widely applied to the technical field of central air conditioners.

The invention discloses an internet online diagnosis and treatment data processing system based on a big data model, and relates to the technical field of medical treatment, an association rule analysis module analyzes association between a disease and drug treatment by adopting a weighted improvement degree, can accurately identify a relatively good matching relationship between the disease and a drug, and improves the diagnosis and treatment efficiency. The online recommendation module dynamically analyzes the recovery effect of the medicine in different time periods based on the dynamic conditional probability by utilizing the candidate medicine pair and combining the real-time treatment condition of the patient, and the mechanism not only ensures that the recommended medicine is highly matched with the current illness state of the patient, but also improves the accuracy of the medicine recommendation; and medicine selection can be adjusted according to feedback data of the patient. The system performs weighting processing on historical data and performs optimization in combination with data sparsity, so that the problem of inaccurate recommendation caused by data sparsity or data insufficiency in a traditional system is effectively avoided.

A movie recommendation method based on collaborative enhancement and graph attention neural network, comprising: first, calculating the respective collaborative movie neighbor sets of users and movies, and obtaining corresponding collaborative interaction embedding vectors; then, mapping the entities in the sets to a movie knowledge graph for propagation to obtain corresponding layer knowledge graph embedding vectors and multi-hop knowledge graph embedding vectors; finally, combining the movie knowledge graph embedding vectors and the collaborative interaction embedding vectors to obtain final user embedding vectors and movie embedding vectors, and calculating the predicted click probability of the user on the movie to obtain the predicted ranking result of the movie recommended to the user from high to low according to the probability. The present application considers the sufficient combination of collaborative information and knowledge graph information, has high accuracy and good recommendation effect.

The present invention provides an item recommendation data processing method, device and storage medium, belonging to the field of data analysis technology, the method comprising: importing a user set, an item positive sample set and an item negative sample set; constructing a training model, training the training model according to the user set, the item positive sample set and the item negative sample set to obtain an item recommendation model; inputting the item negative sample set into the item recommendation model, and obtaining an item recommendation result according to the item recommendation model. The present invention solves the problem of data sparsity, enhances the anti-interference ability of recommendation, and alleviates the problem of data deviation.

The invention discloses a facial expression recognition system and method based on a time sequence and vectorization, and the method comprises the steps: carrying out the processing of an inputted facial video sequence, extracting action units in each frame of image, and arranging the extracted action units into a time sequence according to a time sequence, the method comprises the following steps of: obtaining a plurality of action units, expressing all action units at each time point in a feature vector form, calculating the action units expressed in the feature vector form, generating a complete expression time sequence vector by using the calculated feature vectors, and identifying expressions by using a classification model based on the generated expression time sequence vector. According to the method, the time sequence and the vectorization technology are combined, the problems of sparsity and inaccurate recognition of action units are effectively solved, and the stability and the accuracy of expression recognition are remarkably improved through dynamic time sequence representation.

The invention relates to the technical field of automatic driving, in particular to an open set 3D occupancy prediction framework suitable for an automatic driving scene, which comprises a multi-source data processing module, an occupancy prediction processing module, a model training optimization module and a post-processing and visualization module. According to the open set 3D occupancy prediction framework suitable for the automatic driving scene and the implementation method, through a dynamic-static separation fusion-Poisson reconstruction technology, high-quality dense 3D occupancy representation is generated, holes are filled, semantic differences between a dynamic target and a static scene are reserved, the void rate of a voxel grid after densification is reduced in a nuScenes data set test, and the performance of the open set 3D occupancy prediction framework suitable for the automatic driving scene is improved. The method has the advantages that the accuracy of semantic tags is improved, the problem of sparsity of 3D data is solved, 2D visual features, 3D voxel features and CLIP text features are aligned through knowledge distillation loss, and high-precision prediction of known categories and effective recognition of unknown categories are realized at the same time through double prediction heads and an unknown category decision-making mechanism.

The invention provides an event stream data lane line extraction and detection method based on momentum distillation, and the method comprises the steps: 1, carrying out the preprocessing of sparse event stream data, and determining an event stream representation mode; 2, building a multi-scale event flow feature extractor and a regression module as basic models; 3, copying the basic model as a momentum model; the event flow representation mode obtained in the first step is input into a basic model and a momentum model at the same time, two lane line extraction results are obtained, and the result obtained by the momentum model serves as a pseudo label to supervise training of the basic model together with a real label; only the parameters of the basic model network are updated; 4, updating parameters of the momentum model by adopting an index moving average strategy; and 5, repeating the step 3 and the step 4 until the total loss is converged. According to the invention, the problems of misjudgment, sparsity of event stream data and particularity of lane line shapes are solved.

The present invention discloses a recommendation method based on a knowledge-aware hypergraph neural network, comprising the following steps: step 1, constructing a user hypergraph, with the initial hyperedges taking items that have an interactive relationship with the user as entity nodes; constructing an item hypergraph, with the initial hyperedges taking items that have an interactive relationship with any user that has an interactive relationship with the item as entity nodes; step 2, performing convolution calculations on the user hypergraph and each item hypergraph; step 3, calculating the inner product of the user and all items to obtain the interaction score between the user and all items. It has the beneficial effect of integrating the auxiliary information of the item into the vector representation of the user, so that the items with which the user is more likely to have an interactive relationship can be selected from a large number of items. A recommendation system based on a knowledge-aware hypergraph neural network is disclosed, comprising: a data capture module; a knowledge-aware hypergraph construction module; a domain convolution module; a hyperedge convolution module; and a prediction module. It has the beneficial effects of simple operation and accurate prediction.

The present invention discloses an optimization method and service system for precise push based on big data analysis, which relates to the field of human resources technology. In S1, various job requirements and user resume data are collected and integrated in advance to generate a static data set, which provides basic information support for subsequent matching. S2 uses the job matching factor Gpyz to preliminarily screen users and jobs that have not communicated, so as to accurately locate potential compatible target users and target jobs, thereby improving matching efficiency. In S3, the method generates a dynamic data set and monitors the activity level by counting the behaviors of target users and actively interacting users, analyzes the similarity Xsd, and constructs the satisfaction coefficient of active users for various jobs, further enhancing the understanding of job requirements and user preferences. S4 predicts the satisfaction coefficient and combines the evaluation threshold Q for further comparative analysis, so that the system can further screen out candidates that are highly matched with job requirements.

The present invention discloses a method, system and medium for mining dynamic topic evolution based on a dynamic academic network. The method includes: extracting phrase-level technical words from text information, using the extracted technical words and papers to build a dynamic academic network; obtaining a technical word dynamic vector with time series characteristics based on the dynamic academic network, and generating a technical word distribution of the paper in a global sense based on the academic heterogeneous network of the final time series of the dynamic academic network; inputting the technical word dynamic vector and the technical word distribution into a dynamic topic model, generating the distribution of paper topics at different times, modeling the time series characteristics of paper topics, and obtaining the dynamic evolution trajectory of paper topics at different times. The present invention improves the recognition accuracy of technical words, enhances the modeling ability of the dynamic topic model, and enhances the cross-paper dissemination ability of technical words, which is conducive to generating more interpretable and higher quality topics.

This application provides a source code generation method and apparatus. The method includes: determining target features based on a first fragment, wherein the first fragment includes one or more codes input by a user before a first moment, and the target features are used to indicate code attributes of the codes in the fragment; generating multiple candidate fragments; and selecting a second fragment from the multiple candidate fragments based on the target features, wherein the features of the second fragment have the highest similarity to the target features among the features of the multiple candidate fragments. The target features are obtained based on the source code input by the user and serve as the user's sequence preference. The source code suggestions provided to the user are generated based on the user's sequence preference. In other words, the generated source code is related to the source code input by the user. Therefore, the accuracy of the generated source code can be improved.

The present invention discloses a recommendation method, system and device based on multiple objectives and multiple tasks, the method comprising: obtaining relevant data to form a data set, preprocessing the data set to obtain a processed data set; forming a feature set based on user-related data and financial management-related data, the feature set comprising a user feature set, a financial management-related feature set, a cross feature set, a sequence feature set and an intention feature set; constructing a product recommendation pre-training model; training the product recommendation pre-training model based on the feature set to obtain a product recommendation model; the product recommendation model performs inference prediction on the target-related data to obtain a total probability and then obtain a financial management product recommendation result. The present invention constructs a multi-task and multi-objective product recommendation model by deeply analyzing the complete behavior chain of the user, accurately capturing and expressing the before-and-after dependencies between the behaviors in the chain, and providing users with more accurate behavior predictions.

The invention relates to a method for improving the precision of laser radar point cloud data. The method comprises the following steps: obtaining original low-precision laser radar point cloud data and original high-precision laser radar point cloud data; performing deletion processing on abnormal data in the original low-precision laser radar point cloud data; performing interpolation processing on the low-precision laser radar point cloud data after deletion processing so as to realize matching of the interpolated low-precision laser radar point cloud data and the original high-precision laser radar point cloud data; performing adaptive Fourier decomposition on the interpolated low-precision laser radar point cloud data until the similarity meets a preset condition; the similarity is calculated according to the original high-precision laser radar point cloud data and the interpolated low-precision laser radar point cloud data.

The invention discloses an intention-oriented self-supervised learning graph recommendation method, system and device and a storage medium, and the method comprises the steps: constructing a user-article interaction graph according to the interaction data of a user and an article, inputting the ID of the user-article in the user-article interaction graph into an intention-perceived graph encoder, obtaining a user-article embedded representation under each intention and aggregating the user-article embedded representation; inputting the aggregated user-article embedded representation into an online network in a trained self-supervised learning framework, and predicting a user-article recommendation score; wherein the self-supervised learning framework comprises an online network and a target network, the online network is used for predicting a user-article recommendation score according to the aggregated user-article embedded representation, and the target network is used for performing data enhancement on the aggregated user-article embedded representation in a training process. According to the method, the interest of the user can be better modeled according to the intention, so that more accurate and more efficient user recommendation is realized.

The invention discloses a diversity recommendation optimization method fusing inventory information. The diversity recommendation optimization method comprises the following steps: 1, constructing a data set of users and commodities; 2, constructing a matrix decomposition model; 3, constructing a utility maximization function; embedding commodity inventory constraints and commodity type constraints to optimize recommendation results; 4, a heuristic algorithm is adopted to solve the predicted score of the user for the commodity, and a final recommendation result is obtained. According to the method, the commodity recommendation model based on matrix decomposition is adopted, the inventory information of the commodities is fused, and the situation that the recommendation system recommends out-of-stock products is reduced, so that the recommendation fairness of the system and the satisfaction degree of the user to the platform can be improved.