Content extraction studies using total distance algorithm

With the rapid increase of web usage, it is becoming very difficult to make web pages reach to every region. Especially web pages in foreign languages have fewer complexities when compared with Indian websites. This is because of variations observed in script in Indian context. The present study of this paper focuses on content in the input instead of other data mining and image processing techniques available. Bin-packing strategy, partial digest algorithm and median string algorithms gave good results to identify content. Total distance overcame the drawbacks of hamming distance. The results of median string clearly showed that regional languages in India have greater similarities than English. The paper gives a detailed study of total distance algorithms with obtained results.