Oman Somantri
Department of Informatics, Politeknik Harapan Bersama Tegal

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

Found 2 Documents
Search

Peningkatan Akurasi Klasifikasi Tingkat Penguasaan Materi Bahan Ajar Menggunakan Jaringan Syaraf Tiruan Dan Algoritma Genetika Oman Somantri; Slamet Wiyono
Jurnal Teknologi dan Sistem Komputer Volume 5, Issue 4, Year 2017 (October 2017)
Publisher : Department of Computer Engineering, Engineering Faculty, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (273.84 KB) | DOI: 10.14710/jtsiskom.5.4.2017.147-152

Abstract

Decision support systems can be applied to perform a lecturer's performance assessment. This research aims to develop a hybrid model using the artificial neural network (ANN) and genetic algorithm (GA) that can be implemented and used as a model of decision support data analysis that produce better accuracy, specifically to assess the lecturer's comprehension of their teaching materials. The use of GA in determining the ANN parameter has increased the accuracy from 85.36% to 85.73%. The training cycle is also reduced to 624 from 1000. The use of this JST-GA model can be applied to provide a better lecture's performance assessment system.
Opinion Mining on Culinary Food Customer Satisfaction Using Naïve Bayes Based-on Hybrid Feature Selection Oman Somantri; Dyah Apriliani
Indonesian Journal of Electrical Engineering and Computer Science Vol 15, No 1: July 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v15.i1.pp468-475

Abstract

Conducting an assessment of consumer sentiments taken from social media in assessing a culinary food gives useful information for everyone who wants to get this information especially for migrants and tourists, in th other hand that information is very valuable for food stall and restaurant owners as information in improvinf food quality. Overcoming this problem, a sentiment analysis classification model using naïve bayes algorithm (NB) was applied to get this information. This problem occurs is the level of accuracy of classification of consumer ratings of culinary food is still not optimal because the weight of values in the data preprocessing process are not optimal. In this paper proposed a hybrid feature selection models to overcome the problems in the process of selecting the feature attributes that have not been optimal by using a combination of information gain (IG) and genetic algorithm (GA) algorithms. The result of this research showed that after the experiment and compared to using others algorithms produce the best of the level occuracy is 93%.