Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal EECCIS
Ranap Sitorus
STT Atlas Nusantara Malang
Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analisis Pengaruh Frasa Pada Deteksi Emosi Dari Teks Menggunakan Vector Space Model Ranap Sitorus; Harry Soekotjo Dachlan; Wijono Wijono
Jurnal EECCIS (Electrics, Electronics, Communications, Controls, Informatics, Systems) Vol. 11 No. 1 (2017)
Publisher : Fakultas Teknik, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (592.328 KB) | DOI: 10.21776/jeeccis.v11i1.439

Abstract

Text communication is one way of conveying information about one's attitudes and emotional state. Emotion is also very important role in taking a decision such as doing emotion detection from the text of the questionnaire. Text emotions are detected at various stages to be recognizable by the computer. The steps taken are preprocessing the case folding, tokenizing, filtering and stemming. In this emotional detection research, For grouping words based on word class, using POS-Tagging where approach based rule rule based containing combination of word class which most likely when combined will form a phrase in order to facilitate the computer in understanding the characteristics of a phrase. The phrases-based detection-based tokenisation process uses the Hidden Markov Model (HMM) POS-Tagger for easy classification using the Tf-Idf and VSM methods. This study aims to classify text communication in Indonesia language into emotional expression classes into 3 basic emotional classes and serve as training documents and test documents obtained from the results of student questionnaires as many as 264 documents. The computer is able to detect emotion with the corpus data by performing sentences into words or phrases using Chunk that can classify emotions happy, disappointed and afraid. From the test results for 90% training data and 10% test data in detecting emotions using Phrase got 92.59%
Co-Authors HARRY SOEKOTJO DACHLAN Wijono Wijono