Download now Free registration required
Emergence of data mining methods in data representation has resulted in discovering knowledge from large database systems. Efficient algorithms to mine frequent patterns are crucial to many tasks in spatial association rule mining. A promising approach for mining such spatial frequent pattern is mining frequent sub-patterns which are closed and maximal patterns. In this paper, the authors make use of numerical representations and its arithmetic operations to considerably reduce the size of the transaction dataset. The proposed approach generates a TFP-tree that simplifies the generations of frequent patterns.
- Format: PDF
- Size: 363.41 KB