A Semi-Supervised Self-Organizing Map with Adaptive Local Thresholds
This is an incremental improvement for researchers and practitioners in semi-supervised learning, offering a more robust method for tasks with limited labeled data.
The authors tackled the problem of semi-supervised learning by proposing ALTSS-SOM, a method that dynamically switches learning forms based on label availability and adjusts to local cluster variance, resulting in improved classification performance over other semi-supervised methods and better clustering than pure methods when labels are unavailable, with reduced sensitivity to parameters.
In the recent years, there is a growing interest in semi-supervised learning, since, in many learning tasks, there is a plentiful supply of unlabeled data, but insufficient labeled ones. Hence, Semi-Supervised learning models can benefit from both types of data to improve the obtained performance. Also, it is important to develop methods that are easy to parameterize in a way that is robust to the different characteristics of the data at hand. This article presents a new method based on Self-Organizing Map (SOM) for clustering and classification, called Adaptive Local Thresholds Semi-Supervised Self-Organizing Map (ALTSS-SOM). It can dynamically switch between two forms of learning at training time, according to the availability of labels, as in previous models, and can automatically adjust itself to the local variance observed in each data cluster. The results show that the ALTSS-SOM surpass the performance of other semi-supervised methods in terms of classification, and other pure clustering methods when there are no labels available, being also less sensitive than previous methods to the parameters values.