Probabilistic Semi-Supervised Learning via Sparse Graph Structure Learning

Abstract

We focus on developing a novel scalable graph-based semi-supervised learning (SSL) method for a small number of labeled data and a large amount of unlabeled data. Due to the lack of labeled data and the availability of large-scale unlabeled data, existing SSL methods usually encounter either suboptimal performance because of an improper graph or the high computational complexity of the large-scale optimization problem. In this paper, we propose to address both challenging problems by constructing a proper graph for graph-based SSL methods. Different from existing approaches, we simultaneously learn a small set of vertexes to characterize the high-dense regions of the input data and a graph to depict the relationships among these vertexes. A novel approach is then proposed to construct the graph of the input data from the learned graph of a small number of vertexes with some preferred properties. Without explicitly calculating the constructed graph of inputs, two transductive graph-based SSL approaches are presented with the computational complexity in linear with the number of input data. Extensive experiments on synthetic data and real datasets of varied sizes demonstrate that the proposed method is not only scalable for large-scale data, but also achieve good classification performance, especially for extremely small number of labels.Zitong Wang is with the

An Emotional Analysis Method Based on Multi Model Ensemble Learning

“…1. Li Wang presented a probabilistic semi-supervised learning (SSL) framework based on sparse graph structure learning, and proposed a simple inference approach for the embeddings of unlabeled data based on point estimation and kernel representation, which will get promising results in the setting of SSL compared with many existing methods and significant improvements on small amounts of labeled data [11]. Wang Hong proposed a form of supervised discrete hash algorithm to learn more stable hash codes by learning mutual similarity and using the relationship between different semantic tags [12].…”

“…In addition, the true positive rate TPR and false positive rate FPR indicators are usually introduced to evaluate the sensitivity and misjudgment of the machine learning model [20], which is shown in formula (11).…”

Efficient Sparse Representation for Learning With High-Dimensional Data

“…Sparse representation is an extremely effective method for exploiting the intrinsic structure of high-dimensional data [21], [25], [37], [43], [45]. The goal of sparse representation is to find a compact representation of high-dimensional data by selecting a small subset of a given dictionary to identify low-dimensional structures in high-dimensional data.…”