A distributed file indexing and search system

Abstract

Nowadays many companies are using computers as a resource for their employees to get their job done. In turn, employees are saving the data they have worked on in their own computers. Over time the data being saved on different computers in the company creates chaos; especially when other employees search for data that they know someone else worked on, but they are unable to find it. Employees are losing their time trying to find the data they require. In some cases research may take from minutes to hours. A company cannot sustain such waste of time. These problems were the reason behind the research in solutions that can reduce this waste of time. Investigation into the matter indicated that an information retrieval system was to be used for this scenario. The task of this system is to store the documents in a structure that it is easily queried in order to retrieve information at a later stage. The presented solution uses client-server architecture where the company computers send documents to the server. The server is the central repository for the company documents. Any employee that requires specific data is able to access the solution search engine where his query is sent to the server and a list of document matches is returned. The solution has gone through evaluation against other information retrieval systems. This was required to analyse the behaviour of the system in comparison to others. With the solution presented, the company achieved its goals; that is, to reduce the wasting of time to locate documents and to have a central repository for the company data.