A self-organizing semantic map for information retrieval
SIGIR '91 Proceedings of the 14th annual international ACM SIGIR conference on Research and development in information retrieval
Case-based reasoning
Self-organizing maps
Intelligent Data Analysis: An Introduction
Intelligent Data Analysis: An Introduction
Hi-index | 0.00 |
Kohonen's Self-organizing Map (SOM) is a means for automatically arranging high-dimensional statistical data. The map attempts to represent all the input with optimal accuracy using a restricted set of models or prototypes. The prototypes also become ordered on the map grid so that similar prototypes are close to each other and dissimilar prototypes far from each other. The SOM is useful in clustering, abstraction, and visualization through dimensionality reduction. It has been used in a multitude of application areas ranging form speech recognition to data mining of tests and form robotics to process monitoring. The unsupervised learning scheme of the SOM makes it well suited for applications in which the input data cannot be labeled. A map is ordered and it follows the patterns of the input data in a non-linear but generalizing fashion. All this makes it well suited for data analysis and many areas in developing intelligent systems. In this article, the general principles of using the SOM in data analysis are considered reflecting on the concept of context. An illustrative experiment of data analysis is presented.