“A new approach for user identification in web usage mining preprocessing”
Arvind kumar Dangi Arvind kumar Dangi · IOSR Journal of Computer Engineering · 2013
Web usage mining is a subset of data mining.In order to huge amount of data but the data is less appropriates "quantity and quality" of the web data is opposite to each other this is the main problem.Web data usage Mining is a variation of this field is untapped source of richly offered free textual information.The importance of web data usage mining is mounting along with the immense volumes of data generated in web habitual existence data always arrives in a various, continuous, rapid and time varying flow.Web data usage mining taking out procedures are important in extracting useful streaming on-line sources.As throughout the globe no. of web users are continuously and rapidly growing, it is necessary for the web usage miners to utilize efficient tools in order to discover, extort, clean and assess the desired information.The data pre-processing stage is the most important phase in the preprocessing for investigation of the web user & his usage behavior.To fulfill this requirement the navigations are recorded in web log file as well as the IP address of the website, session of usage & visited web link.In order to improve the performance & quality of data preprocessing in order to identify unique users and user sessions.We propose a new method for web data preprocessing in which it has three phases."In the first phase some websites are selected and by different locations access these website & by applying the (java) tools & methods then find out the IP address of that websites, session usage time & navigations, in the final phase combine them i.e.(web link navigation + IP address of website + session of usage ).This framework helps to investigate the web user usage behavior efficiently.