基于深度学习的高维稀疏数据组合推荐算法

传统方法在对高维稀疏数据进行检测的过程中,受到高维特征扰动的影响,数据误差较大,因此提出一种基于深度学习的高维稀疏数据组合推荐算法。采用相空间重构方法进行高维稀疏数据的特征重构,根据重构结果结合非线性统计序列分析方法进行高维稀疏数据的回归分析和点云结构重组,在此基础上提取高维稀疏数据的组合特征量;依据特征量提取结果采用特征提取技术抽取高维稀疏数据的平均互信息特征量,并结合关联规则挖掘方法进行高维稀疏数据的主成分分析,挖掘高维稀疏数据的相似度属性类别成分,最终采用深度学习方法进行高维稀疏数据组合推荐过程中的自适应寻优,实现高维稀疏数据的组合推荐。仿真结果表明,采用该算法进行高维稀疏数据推荐的属性归类辨识性较好,特征分辨能力较强,提高了数据的检测和识别能力。

A High-dimensional Sparse Data Combination Recommendation Algorithm Based on Deep Learning

In the process of detecting high-dimensional sparse data,the traditional methods are affected by high-dimensional feature perturbation,and the data error is large.Therefore,a high-dimensional sparse data combination recommendation algorithm based on deep learning is proposed.The phase space reconstruction method is used to reconstruct the features of high-dimensional sparse data.Based on the reconstructed results and the non-linear statistical sequence analysis method,the regression analysis of high-dimensional sparse data and the reorganization of point cloud structure are carried out.On this basis,the combined features of high-dimensional sparse data are extracted.According to the results of feature extraction,the average mutual information features of high-dimensional sparse data are extracted by feature extraction technology.Quantity,principal component analysis of high-dimensional sparse data is combined with association rule mining method,and similarity attribute category components of high-dimensional sparse data are mined.Finally,deep learning method is used for self-adaptive optimization in the process of high-dimensional sparse data combination recommendation,which realizes the combination recommendation of high-dimensional sparse data.The simulation shows that the proposed method has better recognition ability for attribute classification and feature resolution of high-dimensional sparse data recommendation,and improves the detection and recognition ability of data.