Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Mantik Penusa JIPN (Journal of Informatics Pelita Nusantara)
Sumarni Adi
Universitas Amikom Yogyakarta
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Perancangan Klasifikasi Tweet Berdasarkan Sentimen Dan Fitur Calon Gubernur DKI Jakarta 2017 Sumarni Adi
Journal Of Informatic Pelita Nusantara Vol 3 No 1 (2018): Journal Of Informatic Pelita Nusantara
Publisher : STMIK Pelita NUsantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (423.922 KB)

Abstract

One of the fastest growing social media users is Twitter, the number of twitter users mentioned continues to increase 300,000 users every day [1]. Twitter users send twitter posts about the facts and opinions of the government products or services they use or express their political, ideological and interest views. Not to mention also send tweet opinions related leaders or influential public figures in this country. With 55 million tweets each day Twitter has a high update rate [1] and is a highly efficient data warehouse for political and social research, so Twitter is a good place to conduct opinion mining or sentiment analysis in classifying the 2017 Jakarta governor candidate .The classification of tweet data is done by analyzing the sentiments on Indonesian tweet opinions by extracting features using Unigram, negation, term Frequency, and TF-IDF (Term Frequency-Invers Document Frequency). Once extracted, the tweet is classified using the Naïve Bayes Classifier (NBC) algorithm.From the results of designing the twitter classification of Indonesian language using Naïve Bayes Classifier algorithm obtained significant difference in value when compared with manual labeling. Positive and neutral sentiments are significant, while negative sentiments are not significant. Keywords: tweet, sentiment, classification, Naïve Bayes Classifier (NBC)
Penerapan Algoritma Dempster Shaferberbasis Android Pada Sistem Pakar Untuk Mendiagnosa Kerusakan Motor Matic Sumarni Adi; Ike Verawati
Jurnal Mantik Penusa Vol. 22 No. 1 (2018): Special Issue
Publisher : Lembaga Penelitian dan Pengabdian (LPPM) STMIK Pelita Nusantara Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (783.637 KB)

Abstract

Expert System is a computer-based system that combines knowledge, facts, and reasoning techniques in solving problems in a particular field like an expert. Expert systems can function as consultants who advise users as well as assistants to experts. One way to overcome and help detect a person's motor motor damage is to create an expert system based on Android as a media for consultation and monitoring so that the user understands what is happening with his motorbike. The Dempster Shafer method is a non monotonous reasoning method used to find inconsistencies due to the addition or subtraction of new facts that will change the existing rules, so the Dempster Shafer method allows one to be safe in doing the work of an expert in this matter is a mechanic. The purpose of this study was to apply the Dempster Shafer uncertainty method to the expert system to diagnose damage to the motorbike and also measure the accuracy of the Dempster Shafer inference engine. The diagnostic results of damage to the matic motor generated by the expert system are the same as the results of manual calculations using the Dempster Shafer inference engine theory which is processed from the results of interviews with the mechanic motor mechanic. So it can be concluded that an android-based expert system that has been built can be used to diagnose damage to the Matic motor. Keywords:Dempster Shafer, Motor Matic, Expert System, Android 
Co-Authors Ike Verawati