Suffix Stripping Problem as an Optimization Problem
Pawan Tamta, Bhagwati Prasad Pande, Hoshiyar S. Dhami · Glottotheory · 2015
Abstract Stemming or suffix stripping is the problem of removing suffixes from words to get the root word. Word endings can be removed by developing stripping rules dependent on the morphological knowledge of a specific language; obviously such approach cannot flourish in multilingual environment. Statistical approaches survive in multilingual environment but they require significant amount of computing. We define stemming as an optimization problem for the very first time in the literature. An Integer Program is being developed for the stemming problem. We exhibit our approach by applying it to clusters of English and Spanish words; moreover, the proposed method is also being compared with an established technique in the field for English language. An AMPL program of the proposed method has also been given in Appendix (A.2).