Abstract:
As there are various pattern matching algorithms are present now but if we try to make them advanced by some new algorithms than it can be made faster easily. In such way the idea is to make a new file and store it in the file system which will be containing the metadata (lengths of words) about the file. Whena system call is made, new associated file will be opened in hidden mode. As this new file is containing metadata about the file mainly containing the lengths of all the words by their position; by which the length of any word in file could be determined. When we create a file then an algorithm works which calculates lengths of all the words present in file and store the lengths in a positional way like first phrase is length of first word and 99th phrase is length of 99th word. This algorithm will work on a file a single time until it is not modified or updated. Once this file has been made after that on a search request it calculates the length of the searched word and finds that length in metadata file. The Positions at which length is matched, only those words will be checked for matching of patterns. So we do not need to compare all the words in file with searched word by which it will reduce its time complexity and searching process will be faster with some storage overhead. The most required thing in computer system is reduced time complexity meanwhile some storage overhead is allowed. Like in binary search firstly we sort the data in ascending order then the binary search is performed so over all reduced complexity is occurred.
Keywords: