Cluster Connections: A visualization technique to reveal cluster boundaries in self-organizing maps
نویسندگان
چکیده
The self-organizing map is one of the most prominent unsupervised learning architectures used to visualize the similarities of high-dimensional input structures. What remains by no means straightforward , is an explicit representation of cluster boundaries in the nal two-dimensional map display. The detection of these boundaries rather requires some amount of insight into the inherent structure of the input data which may not be expected in real-world application scenarios. In this paper we address this deeciency by suggesting an extension to the standard map representation that leads to an easy recognition of cluster boundaries. The general idea is the visualization of clusters within the input data items by connecting units representing similar data items while disconnecting units representing dissimilar data items. As a result we get a grid of connected nodes where the intensity of the connection mirrors the similarity of the underlying data items. Such a representation allows intuitive analysis of the similarities inherent in the input data without the necessity of substantial prior knowledge, and an intuitive recognition of cluster boundaries.
منابع مشابه
Performance Comparison of Particle Swarm Optimization with Traditional Clustering Algorithms used in Self-Organizing Map
Self-organizing map (SOM) is a well known data reduction technique used in data mining. It can reveal structure in data sets through data visualization that is otherwise hard to detect from raw data alone. However, interpretation through visual inspection is prone to errors and can be very tedious. There are several techniques for the automatic detection of clusters of code vectors found by SOM...
متن کاملPerformance Comparison of Particle Swarm Optimization with Traditional Clustering Algorithms used in Self-Organizing Map
Self-organizing map (SOM) is a well known data reduction technique used in data mining. It can reveal structure in data sets through data visualization that is otherwise hard to detect from raw data alone. However, interpretation through visual inspection is prone to errors and can be very tedious. There are several techniques for the automatic detection of clusters of code vectors found by SOM...
متن کاملAlternative Ways for Cluster Visualization in Self-Organizing Maps
We present two enhanced visualization techniques for the self-organizing map allowing the intuitive representation of input data similarity. The general idea of both approaches is to visualize the relationship of nodes to facilitate the detection of cluster boundaries without modifying the architecture or the basic training process of SOM. One approach mirrors the movement of weight vectors dur...
متن کاملAbstract—self-organizing Map (som) Is a Well Known Data
reduction technique used in data mining. It can reveal structure in data sets through data visualization that is otherwise hard to detect from raw data alone. However, interpretation through visual inspection is prone to errors and can be very tedious. There are several techniques for the automatic detection of clusters of code vectors found by SOM, but they generally do not take into account t...
متن کاملFree Projection SOM: A New Method For SOM-Based Cluster Visualization
In this paper an extension to the learning rule of the Self-Organizing Map (SOM) namely the Free Projection SOM (FP-SOM) is presented in order to enhance the SOM projection. The general idea of the FPSOM is to mirror the movement of weight vectors during the training process allowing their images on the map grid to move more freely between the junctions. The result of the extended training algo...
متن کامل