ABSTRACT
We focus on the problem of finding patterns across two large, multidimensional datasets. For example, given feature vectors of healthy and of non-healthy patients, we want to answer the following questions: Are the two clouds of points separable? What is the smallest/largest pair-wise distance across the two datasets? Which of the two clouds does a new point (feature vector) come from?We propose a new tool, the tri-plot, and its generalization, the pq-plot, which help us answer the above questions. We provide a set of rules on how to interpret a tri-plot, and we apply these rules on synthetic and real datasets. We also show how to use our tool for classification, when traditional methods (nearest neighbor, classification trees) may fail.
- 1.R. Agrawal, J. Gherke, D. Gunopoulos, and P. Raghavan. Automatic subspace clustering of high dimensional data for data mining applications, 1998.Google Scholar
- 2.D. Barbare and P. Chen. Using the fractal dimension to cluster datasets. In Proc. of the 6th International Conference on Knowledge Discovery and Data Mining (KDD-200), pages 260-264, 2000. Google ScholarDigital Library
- 3.A. Belussi and C. Faloutsos. Estimating the selectivity of spatial queries using the 'correlation' fractal dimension. In Proc. of VLDB - Very Large Data Bases, pages 299-310, 1995. Google ScholarDigital Library
- 4.S. Berchtold, C. B6hm, D. A. Keim, and H.-P. Kriegel. A cost model for nearest neighbor search in high-dimensional data space. In Proc. of the I6th A CM SIGA CT-SIGMOD-SIGART Symposium on Principles of Database Systems, pages 78-86, 1997. Google ScholarDigital Library
- 5.S. Berchtold, C. BShm, and H.-P. Kriegel. The pyramid-tree: Breaking the curse of dimensionality. In Proc. A CM SIGMOD Conf. on Management of Data, pages 142-153, 1998. Google ScholarDigital Library
- 6.S. Chaudhuri. Data mining and database systems: Where is the intersection? Data Engineering Bulletin, 21(1):4-8, 1998.Google Scholar
- 7.M.-S. Chen, J. Han, and P. S. Yu. Data mining - an overview from a database perspective. IEEE Transactions on Knowledge and Data Engineering, 8(6):866-883, 1996. Google ScholarDigital Library
- 8.M. Ester, A. Frommelt, H.-P. Kriegel, and J. Sander. Algorithms for characterization and trend detection in spatial databases. In Proc. of the 4th International Conference on Knowledge Discovery and Data Mining (KDD-98), pages 44-50, 1998.Google Scholar
- 9.M. Ester, H.-P. Kriegel, and J. Sander. Spatial data mining: A database approach. In LNCS 1262: Proc. of 5th Intl. Symposium on Spatial Databases (SSD'97), pages 47-66, 1999. Google ScholarDigital Library
- 10.C. Faloutsos, B. Seeger, C. Traina Jr., and A. Traina. Spatial join selectivity using power laws. In Proc. ACM SIGMOD 2000 Conf. on Management of Data, pages 177-188, 2000. Google ScholarDigital Library
- 11.U. M. Fayyad. Mining databases - towards algorithms for knowledge discovery. Data Engineering Bulletin, 21(1):39-48, 1998.Google Scholar
- 12.U. M. Fayyad, C. Reina, and P. S. Bradley. Initialization of iterative refinement clustering algorithms. In Proc. of the 4th International Conference on Knowledge Discovery and Data Mining (KDD-98), pages 194-198, 1998.Google Scholar
- 13.V. Ganti, J. Gehrke, and R. Ramakrishnan. Mining very large databases. IEEE Computer, 32(8):38-45, 1999. Google ScholarDigital Library
- 14.J. Han, L. V. S. Lakshmanan, and R. T. Ng. Constraint-based multidimensional data mining. IEEE Computer, 32(8):46-50, 1999. Google ScholarDigital Library
- 15.C. Traina Jr., A. Traina, L. Wu, and C. Faloutsos. Fast feature selection using the fractal dimension. In XV Brazilian Symposium on Databases (SBBD), 2000.Google Scholar
- 16.R. J. Bayardo Jr., R. Agrawal, and D. Gunopulos. Constraint-based rule mining in large, dense databases. In Proc. of IEEE Intl. Conference on Data Engineering, pages 188-197, 1999. Google ScholarDigital Library
- 17.D. A. Keim and H.-P. Kriegel. Possibilities and limits in visualizing large amounts of multidimensional data. In Perceptual Issues in Visualization. Springer, 1994.Google Scholar
- 18.E. M. Knorr and R. T. Ng. Finding aggregate proximity relationships and commonalities in spatial data mining. IEEE Transactions on Knowledge and Data Engineering, 8(6):884-897, 1996. Google ScholarDigital Library
- 19.R. T. Ng and J. Han. Efficient and effective clustering methods for spatial data mining. In Proc. of VLDB - Very Large Data Bases, pages 144-155, 1994. Google ScholarDigital Library
- 20.Bureau of Census. Tiger/line preeensus files: 1990 technical documentation. Bureau of the Census. Washington, DC, 1989.Google Scholar
- 21.B.-U. Pagel, F. Korn, and C. Faloutsos. Deflating the dimensionality curse using multiple fractal dimensions. In Proc. of IEEE Intl. Conference on Data Engineering, pages 589-598, 2000. Google ScholarDigital Library
- 22.M. Schroeder. Fractals, Chaos, Power Laws. W.H. Freeman and Company, New York, 1991.Google Scholar
- 23.H. G. Schuster. Deterministic Chaos. VCH Publisher, Weinheim, Basel, Cambridge, New York, 1988.Google Scholar
- 24.G. Sheikholeslami, S. Chatterjee, and A. Zhang. Wavecluster: A multi-resolution clustering approach for very large spatial databases. In Proe. of VLDB - Very Large Data Bases, pages 428-439, 1998. Google ScholarDigital Library
- 25.M. L. Tian Zhang, R. Ramakrishnan. Birch: An efficient data clustering method for very large databases. In Proe. of A CM SIGMOD Conf. on Management of Data, pages 103-114, 1996. Google ScholarDigital Library
- 26.R. Weber, H.-J. Schek, and S. Blott. A quantitative analysis and performance study for similarity-search methods in high-dimensional spaces. In Proc. of VLDB - Very Large Data Bases, pages 194-205, 1998. Google ScholarDigital Library
Index Terms
- Tri-plots: scalable tools for multidimensional data mining
Recommendations
Tri-partition neighborhood covering reduction for robust classification
Neighborhood Covering Reduction extracts rules for classification through formulating the covering of data space with neighborhoods. The covering of neighborhoods is constructed based on distance measure and strictly constrained to be homogeneous. ...
Multi-Classification by Using Tri-Class SVM
The standard form for dealing with multi-class classification problems when bi-classifiers are used is to consider a two-phase (decomposition, reconstruction) training scheme. The most popular decomposition procedures are pairwise coupling (one versus ...
A Novel Semi-Supervised SVM Based on Tri-Training
IITA '08: Proceedings of the 2008 Second International Symposium on Intelligent Information Technology Application - Volume 03One of the main difficulties in machine learning is how to solve large-scale problems effectively, and the labeled data are limited and fairly expensive to obtain. In this paper a new semi-supervised SVM algorithm is proposed. It applies tri-training to ...
Comments