Knowledge engineering perspective of text compression

The paper focuses on an engineering perspective of Data Mining, specifically using it as a tool for efficient data compression. Huffman encoding, a lossless compression technique is refined to incorporate frequent itemset mining, an important phase of Association Rule Mining. The research exploits the principle of assigning shorter codes to frequently occurring patterns(sequence of characters) in relation to single character based code assignment approach of Huffman encoding. Simulation results indicate the benefits of the Data Mining approach to compression, resulting in an efficient data compression algorithm.