Effective ranked conceptual retrieval
Amjad M. Daoud · 2007
We describe the design and implementation of the SMARTER prototype system for collecting, structuring and searching information derived from full text ebooks available on the World Wide Web [4]. Standard methods of parsing and extended Boolean schemes for weighting are used to build an index of individual pages. Ranked pages are retrieved from the system by means of a correlated and combined p-norm and vector query mechanisms. Furthermore, we present a complete evaluation of the relative effectiveness of the SMARTER system versus the Google Desktop [7].