Implementation of Porter Stemmer Algorithm to Obtain Basic Words in Toba Batak Language Documents with the Two-Level Morphological Method. (October 2019)

• Record Type: Journal Article
• Authors: Muchtar, M A; Nababan, Erna B; Nababan, Manguji; Andayani, U; Simanjuntak, Tuti; Sitompul, Opim S
• Is Part Of: IOP conference series. Volume 648(2019)
• Publication Date: 2019-10
• Subjects: Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/648/1/012025
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: Stemming is the process of deciphering words by carrying out affixes to words to get the right base words. In its development stemming has been developed for various languages, including Porter Stemmer. In this study, the stem was used to add Toba Batak language. Decapitation of affixes in the Toba Batak language is difficult because there are essential words that contain affixes and there are essential words that decay when affixed, namely the essential words beginning with the letter's 't', 's' and 'p'. To solve this problem by offering the implementation of a porter stemmer algorithm using the two-level morphological method. This stemmer method uses a dictionary of essential words. Each step in the stemmer process will always check the lexicon of the essential words so that the essential words that contain affixes will not be monitored, and if the word is not in the dictionary the basic word will not be stemming so when there are a name people or place names that contain affixes will not be implemented. The use of two levels of morphology was applied to the essential words beginning with the letter's 't', 's' and 'p' to get the right essential words. From the results of testing on 20 Toba Batak language documents, the Porter stemmer algorithm with two levels of morphology can produce an accuracy rate of 87.503% with an average speed of 0.000460798 seconds per word.