Document similarity analysis in Slovak language
Vladimir Hanusniak, Vladimir Smatanik, Milan Straka, Michal Zábovský · 2016
Examining data for similar items is one of the fundamental data-mining problems. Application of methods for similarity search could be useful for plagiarism or near-duplicate web page detection. The computerized methods developed during last years are mainly focused on English language. However, Slovak language has several specific attributes and using these methods may not be precise enough. Our objective of this research paper is to develop a Proof-of-Concept for document similarity estimation process devoted to Slovak Language. The complexity of Slovak language gives us an opportunity to analyze and adjust the methods parameters and thus achieve higher accuracy. Text mining process suggested in this article utilizes stop words list and shingles to accurately measure documents similarity. Results are further validated using date constraint.