This paper presents the Self-Sorting Map (SSM), a novel algorithm for organizing and visualizing data. Given a set of data items and a dissimilarity measure between each pair of them, the SSM places each item into a unique cell of a structured layout, where the most related items are placed together and the unrelated ones are spread apart. The algorithm nicely integrates ideas from dimension reduction techniques, sorting algorithms, and data clustering approaches. Instead of solving the continuous optimizing problem as other dimension reduction approaches do, the SSM transforms it into a discrete labeling problem. As a result, it can organize a set of data into a structured layout without overlapping, providing a simple and intuitive presentation. Experiments on different types of data show that the SSM can be applied to a variety of applications, ranging from visualizing semantic relatedness between articles to organizing image search results based on visual similarities. Our current SSM implementation using Java is fast enough for interactively organizing datasets with hundreds of entries.
展开▼
