The Community for Technology Leaders
Green Image
Issue No. 09 - September (2005 vol. 17)
ISSN: 1041-4347
pp: 1208-1222
Qi Li , IEEE
Hui Xiong , IEEE
Jieping Ye , IEEE
Dimension reduction is a critical data preprocessing step for many database and data mining applications, such as efficient storage and retrieval of high-dimensional data. In the literature, a well-known dimension reduction algorithm is Linear Discriminant Analysis (LDA). The common aspect of previously proposed LDA-based algorithms is the use of Singular Value Decomposition (SVD). Due to the difficulty of designing an incremental solution for the eigenvalue problem on the product of scatter matrices in LDA, there has been little work on designing incremental LDA algorithms that can efficiently incorporate new data items as they become available. In this paper, we propose an LDA-based incremental dimension reduction algorithm, called IDR/QR, which applies QR Decomposition rather than SVD. Unlike other LDA-based algorithms, this algorithm does not require the whole data matrix in main memory. This is desirable for large data sets. More importantly, with the insertion of new data items, the IDR/QR algorithm can constrain the computational cost by applying efficient QR-updating techniques. Finally, we evaluate the effectiveness of the IDR/QR algorithm in terms of classification error rate on the reduced dimensional space. Our experiments on several real-world data sets reveal that the classification error rate achieved by the IDR/QR algorithm is very close to the best possible one achieved by other LDA-based algorithms. However, the IDR/QR algorithm has much less computational cost, especially when new data items are inserted dynamically.
Index Terms- Dimension reduction, linear discriminant analysis, incremental learning, QR Decomposition, Singular Value Decomposition (SVD).
Ravi Janardan, Qi Li, Haesun Park, Vipin Kumar, Hui Xiong, Jieping Ye, "IDR/QR: An Incremental Dimension Reduction Algorithm via QR Decomposition", IEEE Transactions on Knowledge & Data Engineering, vol. 17, no. , pp. 1208-1222, September 2005, doi:10.1109/TKDE.2005.148
84 ms
(Ver )