Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JOIN (Jurnal Online Informatika) Journal of Applied Data Sciences
Buslim, Nurhayati
Unknown Affiliation
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 
Ensemble learning techniques to improve the accuracy of predictive model performance in the scholarship selection process Buslim, Nurhayati; Zulfiandri, Zulfiandri; KyungOh, Lee
Journal of Applied Data Sciences Vol 4, No 3: SEPTEMBER 2023
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v4i3.112

Abstract

Ensemble Learning is an algorithm that searches for the best prediction result based on several classifier solutions which are come from different algorithms. Ensemble learning has better accuracy and performance compared to other algorithms because this method uses several learning algorithms to achieve better predictive solutions. There are a lot of data that the scholarship organizer receives and manages. This makes the process take a lot of time to do checking process and makes it inefficient. Accelerating the process while also maintaining the accuracy of the scholarship selection process can certainly improve the selection performance. In this study, we process student data from UIN Jakarta University as a sample. The model will have 2 base classifiers, namely Support Vector Machine (SVM) and Key Nearest Neighbor (KNN). Each of these algorithms already has quite a good accuracy. Using Ensemble Learning improves the model performance because it has the ability to overcome errors that occur in each data prediction. We can exploit the classification application created using "Streamlit" and will determine whether a student is accepted or rejected in the scholarship selection process. Our model and application can be used by academics as a Decision Support System (DSS) for determining scholarship recipients. This model can also be used as a development for institutions in the academic field.
Analisa Sentimen Menggunakan Data Twitter, Flume, Hive Pada Hadoop dan Java Untuk Deteksi Kemacetan di Jakarta Buslim, Nurhayati; Busman, Busman; Sinatrya, Nadika Sigit; Kania, Tifani Shallynda
JOIN (Jurnal Online Informatika) Vol 3 No 1 (2018)
Publisher : Department of Informatics, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/join.v3i1.141

Abstract

Berikut adalah paragraf yang sudah dirapikan: Traffic congestion in big cities in Indonesia is unavoidable, especially in Jakarta. The increasing number of vehicles and the lack of public transportation are the main causes of traffic congestion in Jakarta. It disturbs people's activities. The government has already made various efforts to resolve the congestion problem; however, it requires high installation and maintenance costs and takes time to be implemented. People often complain about traffic congestion in Jakarta by posting on Twitter, which are called tweets. Every tweet posted is saved in the Twitter API and used for sentiment analysis. It analyzes the emotions of the users. Based on these problems, we conducted research on how to detect traffic congestion in Jakarta. Therefore, we tried to create a Congestion Detection App. We designed the app using UML diagrams. The Congestion Detection App is connected with Hadoop, Flume, Hive, and Derby. The app streams Twitter data collected by connecting with the Twitter API. This app is a Java-based application that can create and view data tables. It performs searches on tweets by ID and analyzes traffic conditions in specific regions in Jakarta. The app performs sentiment analysis on certain tweets and displays the results based on the data tables. The result of the research compares data from the Congestion Detection App with data from Google Maps. We made three value categories consisting of three colors: green for less traffic congestion with a value of 1, orange for medium-scale traffic congestion with a value of 2, and red for heavy traffic congestion with a value of 3. Based on these three categories and values, we used four regions as samples and compared the values with those from Google Maps data to determine accuracy. We achieved an 81% average accuracy from the four samples. The data from the tweet samples compared with Google Maps data showed significant congestion detection with the Congestion Detection App.
Co-Authors Busman Busman, Busman Kania, Tifani Shallynda KyungOh, Lee Sinatrya, Nadika Sigit Zulfiandri Zulfiandri