Title :
Patterns in large numerical data
Author_Institution :
Dept. of Comput. Sci., San Jose State Univ., CA, USA
Abstract :
For time series data, the interest is in "vertical" patterns, not "horizontal" associations; in other words, the focus is on patterns of a large (long) numerical sequence (of vectors or numbers). This paper is theoretical; all data has no noise. It searches for several important mathematical concepts in data mining, such as pattern and prediction and the notion of large. It proposes that data is large if the complexity of data is more than the complexity of the pattern, and reconfirms the previous proposal that a pattern\´s complexity should be smaller than data complexity.
Keywords :
data mining; database theory; sequences; time series; very large databases; complexity; data mining; large numerical data; numbers; patterns; prediction; time series data; vectors; Algebra; Automation; Binary sequences; Computer science; Data mining; Databases; Information theory; Machine learning; Mathematics; Polynomials;
Conference_Titel :
Fuzzy Information Processing Society, 2002. Proceedings. NAFIPS. 2002 Annual Meeting of the North American
Print_ISBN :
0-7803-7461-4
DOI :
10.1109/NAFIPS.2002.1018075
