Enhancing the Productivity of Digital Data Retrieval

Abstract

The escalate of crime rate around the global manifests the lag in the ongoing methodology in extraction or inspection of information retrieved and stored in cache from divers communication channels of the digital investigation system. The Principle objective for the development of this strategy, is to revamp the existing methodology and develop a crime relates information mining framework for extracting and examine pertinent information from stored data which incorporates emails, chat threads and any text messages to discover the criminal activity and solve the enigma with the help of the certainty concealed within the data. This strategy is achieved by the prominence of the three aspects initiated to the input data of textual evidences such as mails, text messages, chat threads, etc. the three aspects are, the separation of the body and header part of a textual corpuses consisting of all kinds of textual evidences of diverse communication channels and preprocessing of text , which is achieved using regular expressions of PHP script and stemming process respectively, in order to make the mining process of text more reliable for the forensic department of crime investigation. Further the searching technique used in this methodology which is constructed for the most highly effective and efficient retrieval of data. Although the technique followed in this methodology is an ancient technique of separation of body and header in a mail, the main aspect of this methodology is to focus over the efficiency in searching technique built for the purpose of effectiveness. A Clustering algorithm is used in this methodology to improvise the system. This algorithm has mainly three feature, it is the alternative form of the Reverse Factor algorithm, it uses bit-parallelism simulation of the suffix automaton of xR and it is efficiency is high if the pattern length is not longer than the memory-word size of the machine. Using this kind of technique to improvise the existing system would bring about a methodical procedure, which would initiate in a highly efficacious searching system, evidenced by the time complexity, precision and recall value.