EVALUASI EFEKTIVITAS PENCARIAN DOKUMEN HTML PADA PENERAPAN STEMMED TERM VECTOR MODEL DENGAN PEMBOBOTAN LOGARITMA FREKUENSI TERM DALAM DOKUMEN DAN PENGUKURAN SOKAL

Abstract

Purpose of this study was to evaluate the effectiveness of an information retrieval system using the sokal / sneath equation and vector method by comparing between one system model and another. Number of documents used 100 documents taken from the internet with 10 different topics. Algorithms used include algorithm for extact, tokenization, stopword removal, stemming, term weighting and similarity calculation. An information acquisition system is a system that automatically searches for information that is relevant to user needs. Finally, in measuring the relevance of the document to the query is done using the calculation of the average precision and recall. The calculations are done manually. From the calculation of the average precision and recall will get a collection point of effectiveness later on at the end of this study will be described in graphical form. Furthermore, the graph is incorporated into several graphs of precision and recall in other related studies, for comparison in obtaining final conclusions..