Automated classification of web pages in hierarchical browsing

摘要

Purpose - The purpose of this study is twofold: to investigate whether it is meaningful to use thernEngineering Index (Ei) classification scheme for browsing, and then, if proven useful, to investigate thernperformance of an automated classification algorithm based on the Ei classification scheme.rnDesign/methodology/approach - A user study was conducted in which users solved fourrncontrolled searching tasks. The users browsed the Ei classification scheme in order to examine thernsuitability of the classification systems for browsing. The classification algorithm was evaluated byrnthe users who judged the correctness of the automatically assigned classes.rnFindings - The study showed that the Ei classification scheme is suited for browsing. Automaticallyrnassigned classes were on average partly correct, with some classes working better than others. Successrnof browsing showed to be correlated and dependent on classification correctness.rnResearch limitations/implications - Further research should address problems of disparaternevaluations of one and the same web page. Additional reasons behind browsing failures in the Eirnclassification scheme also need further investigation.rnPractical implications - Improvements for browsing were identified; describing class captionsrnand/or listing their subclasses from start; allowing for searching for words from class captions withrnsynonym search (easily provided for Ei since the classes are mapped to thesauri terms); whenrnsearching for class captions, returning the hierarchical tree expanded around the class in whichrncaption the search term is found. The need for improvements of classification schemes was alsornindicated.rnOriginality/value - A user-based evaluation of automated subject classification in the context ofrnbrowsing has not been conducted before; hence the study also presents new findings concerningrnmethodology.