2019 : Comparison of stemming algorithms on Indonesian text processing

Dr. Ir. Aris Tjahyanto M.Kom.


Abstract

[...] is to identify the best algorithm for Indonesian text processing purpose. If words have the same roots, they are considered to have a semblance of meaning.[...] documents that have words with the same roots are considered relevant so the stemming process will reduce the features dimension of the documents [4].[...] this does not occur in porter algorithm. Besides the stemmer errors that occur in the porter algorithm, it was also due to the rules which did not match the Indonesian language. 4.3. Correlation Test Result between Stemmer Performance and Clustering Performance This section discusses the results of correlation testing between the stemmer algorithm performances and clustering performance using Pearson method.[...] International Conference on Emerging Trends in Engineering and Technology.