Clustering method based on comparative learning, system and equipment and storage medium

Abstract

The invention relates to artificial intelligence, and provides a clustering method based on comparative learning, and the method comprises the steps: obtaining sample data which comprises label data; obtaining a first similar label matrix and a first cosine similarity matrix according to the sample data; obtaining a first loss function according to the first similar label matrix and the first cosine similarity matrix; training according to the first loss function and the sample data to obtain an optimization model; respectively processing the label data and the sample data based on an optimization model to obtain a first centroid sequence and a second centroid sequence; according to the first centroid sequence and the second centroid sequence, performing label labeling processing on the sample data to obtain pseudo label data; training the optimization model according to the pseudo label data to obtain a clustering model; and inputting the sample data into the clustering model to obtain a clustering result. According to the invention, hybrid clustering can be carried out by using the label data and the label-free data, the accuracy of data category identification is improved, and the clustering effect is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a clustering method, system, device and storage medium based on contrastive learning. Background technique [0002] Cluster analysis refers to the analytical process of grouping a collection of data objects into multiple classes composed of similar objects. In related technologies, data with similar tags are analyzed and identified according to tags carried by the data, and data with similar tags are grouped into a group, thereby being divided into different groups. The data will be divided into different groups due to its own label, so the clustering effect will be affected by the accuracy of the label. However, accurate label data is usually manually labeled, and it is difficult to obtain enough data. The cost of clustering analysis using only label data is too high, and the clustering results are unstable and the clustering effect is poor due to insuff...

AI Technical Summary

Problems solved by technology

However, accurate label data is usually manually labeled, and it is difficult to obtain enough data. The cost of clustering analysis using only label data is too high, and the clustering results are unstable due to insufficient data volume, and the clustering effect is poor.

However, using unlabeled data with only features and no labels for cluster analysis is prone to cluster point offset, unstable results, and poor clustering effect

Images