Database-friendly random projections
PODS '01 Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Random projection in dimensionality reduction: applications to image and text data
Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
Self-Organizing Maps
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Improved heterogeneous distance functions
Journal of Artificial Intelligence Research
Random projection, margins, kernels, and feature-selection
SLSFS'05 Proceedings of the 2005 international conference on Subspace, Latent Structure and Feature Selection
| Hi-index | 0.00 |
Visualisation of multi-dimensional dataset can be very useful for data mining purposes. This paper describes a simple visualisation technique to reduce a high dimensional dataset into a 3D space. Our aim is to design a method that is simple, easy, computationally cost-effective and able to give a reasonable visualisation of the dataset. The original dataset is projected to a lower dimensional space via geometric metrics, while the proximity of the original data points is approximately preserved. The idea behind our data transformation is the concept of triangulation, which is applied through the use of reference points. In our study, we compared our method with the Principal Component Analysis (PCA) and Random Projection (RP). The results suggest that: when compared with PCA, our method can deliver a comparable visualisation of the dataset at a lower cost; when compared with RP, our method yields better visualisations at a similar cost.