Multimodal heterogeneous big data classification optimization method based on Naive Bayes
Due to the high dimension of multimodal heterogeneous big data,the spatial distribution of classification modes is inconsistent with the actual distribution,and the accuracy of data classification is low.To solve this problem,a multimodal heterogeneous big data classification optimization method based on Naive Bayes is proposed.Utilize TF-IDF algorithm to eliminate data with low resource utilization.Naive Bayes classification algorithm is used to obtain the probability of data occurrence and classify big data.Dimension reduction deals with multi-source heterogeneous big data,and classifies data through Naive Bayes decision-making.The experimental results show that the spatial distribution of the studied method's modality is only not completely consistent with the actual distribution of modality 3,and the rest are consistent.The classification accuracy ultimately reaches 0.93%,indicating a precise classification effect.
Naive Bayesmultimodal heterogeneousbig data classificationTF-IDF algorithm
万方数据
维普
