Claim Missing Document
Check
Articles

Found 8 Documents
Search
Journal : The Indonesian Journal of Computer Science

Sistem Pendukung Keputusan Kelayakan Sertifikasi Guru Menggunakan Metode Multi Attribute Utility Theory (MAUT) Pada SMAN 2 Mandau Harianto, Helfin; Agustin; Junadhi; Tashid
The Indonesian Journal of Computer Science Vol. 12 No. 2 (2023): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v12i2.3169

Abstract

Pelaksanaan sertifikasi guru merupakan komitmen pemerintah untuk mengimplementasikan amanat Undang-Undang Nomor 14 tahun 2005, yakni mewujudkan guru yang berkualitas dan profesional. Hal-hal terkait dengan proses sertifikasi masih belum sepenuhnya menggunakan sistem yaitu masih dengan cara mendata guru yang layak mengikuti proses sertifikasi berdasarkan kriteria masa kerja, usia, pendidikan terakhir, tugas tambahan, prestasi mengajar, dan jumlah jam mengajar sehingga sering kali menimbulkan kesulitan ketika mengusulkan guru yang layak mengikuti proses sertifikasi dikarenakan memakan waktu yang lama saat pengurutan ranking sertifikasi. Banyak guru yang mengeluhkan proses sertifikasi yang tidak transparan, diantaranya guru yang usia muda serta masa kerja yang lebih sedikit mendapat kesempatan lebih dulu menjalani proses sertifikasi daripada guru yang sudah mempunyai pengalaman kerja yang lama dan usia tua. Sistem pendukung keputusan merupakan proses tindakan atau aksi dalam pemecahan masalah yang diyakini akan memberikan solusi terbaik untuk mencapai tujuan. Dalam Sistem Pendukung Keputusan ini digunakan metode Multi Attribute Ultility Theory (MAUT). Hasil penelitian menggunakan metode mampu memberikan rekomendasi guru yang layak mengikuti sertifikasi. Penerapan metode MAUT memberikan hasil akurasi sebesar 90%, dari hasil tersebut menunjukkan bahwa metode MAUT bisa menjadi metode alternatif untuk sistem kelayakan sertifikasi guru.
Perbandingan Algoritma K-Means dan K-Medoids untuk Pengelompokan Program BPJS Ketenagakerjaan Meiriza, Adellia; Ali, Edwar; Rahmiati; Agustin
The Indonesian Journal of Computer Science Vol. 12 No. 2 (2023): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v12i2.3184

Abstract

BPJS Ketenagakerjaan bertugas menyelenggarakan program jaminan sosial bagi para pekerja di Indonesia, seperti Jaminan Kecelakaan Kerja, Jaminan Hari Tua, Jaminan Pensiun, Jaminan Kematian, dan Jaminan Pemeliharaan Kesehatan. Pengelompokan program bukan penerima upah dapat menggunakan metode clustering. Dalam penelitian ini, peneliti membandingkan dua algoritma clustering yaitu K-Means dan K-Medoids untuk mengelompokkan program bukan penerima upah berdasarkan karakteristik yang dimiliki. Data yang digunakan dalam penelitian ini diperoleh dari BPJS Ketenagakerjaan cabang pekanbaru. Pengelompokan dilakukan dengan menggunakan jumlah cluster yang sama untuk kedua algoritma yaitu K = 3. Hasil dari penelitian menunjukkan bahwa K-Medoids menghasilkan kelompok yang lebih stabil dan robust dibandingkan dengan K-Means. Hasil nilai DBI menunjukkan bahwa K-Medoid lebih baik dari K-Means. Hasil ini dapat dijadikan rekomendasi kepada pendaftar yang akan mengambil program BPJS Ketenagakerjaan selain itu penggunaan K-Medoids sebagai algoritma clustering lebih efektif dibandingkan K-Means untuk pengelompokan program bukan penerima upah.
Optimization of Deep Learning with FastText for Sentiment Analysis of the SIREKAP 2024 Application Handoko; Junadhi; Triyani Arita Fitri; Agustin
The Indonesian Journal of Computer Science Vol. 14 No. 2 (2025): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v14i2.4809

Abstract

This study analyzes public sentiment towards the SIREKAP 2024 application using deep learning. Data was collected from Google Playstore reviews and processed through cleaning, tokenization, and stemming. Word embedding was performed using FastText to capture more accurate word representations, including OOV words. The deep learning models compared were CNN, BiLSTM, and BiGRU. Performance evaluation used accuracy, precision, recall, and F1-score metrics. The results showed that the CNN model with FastText Gensim embedding achieved the highest accuracy of 95.98%, outperforming BiLSTM and BiGRU. This model was more effective in classifying positive and negative sentiments. This study provides insights for developers to improve the performance and public trust in SIREKAP 2024 and opens opportunities for further research with more complex embedding approaches and deep learning models.
Analisis Pilkada Medan pada Sosial Media Menggunakan Analisis Sentimen dan Social Network Analyisis Anam, M. Khairul; Firdaus, Muhammad Bambang; Fitri, Triyani Arita; Lusiana; Agustin, Wirta; Agustin
The Indonesian Journal of Computer Science Vol. 11 No. 1 (2022): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v11i1.3027

Abstract

The simultaneous regional head elections were over, but during the campaign until it was decided to become regional head there were many comments, both pro and contra. The city of Medan is one of the regions that will hold the 2020 ELECTION during the pandemic. The Medan City Election has decided that the pair Bobby Nasution and Aulia Rachman have won. This victory certainly gets a variety of comments on social media, especially Twitter. This study conducts sentiment analysis to see the sentiment that occurs, namely seeing negative, positive, or neutral comments. This sentiment analysis uses two methods to see the resulting accuracy, namely Support Vector Machine (SVM) and Naïve Bayes Classifier (NBC). This study also looks at the interactions that occur using Social Network Analysis (SNA). In addition to sentiment analysis and SNA, this study also looks at the existence of BOT accounts used in the #PilkadaMedan. The results obtained from the sentiment analysis show that NBC has the highest accuracy, which is 81, 72% with a data proportion of 90:10. Then on SNA, the @YanHarahap account got the highest nodes, namely 911 nodes. Then from 10326 tweets, 11% were suspected of being BOT by the DroneEmprit Academic system.
Analisis Sentimen Layanan Hotel Menggunakan Algoritma Extra Trees: Studi Kasus pada Ulasan Pelanggan Aprilita, Windi; Junadhi; Agustin; Hadi Asnal
The Indonesian Journal of Computer Science Vol. 13 No. 3 (2024): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i3.4014

Abstract

This research aims to analyze the sentiment of hotel services based on customer reviews using the Extra Trees algorithm. This method was tested on a dataset containing customer reviews about hotel services. The evaluation is done by taking into account the accuracy, precision, recall, and F1 score of the developed model. The results showed that the Extra Trees algorithm was able to achieve an accuracy of 85.05%, with a precision of 84.46%, a recall of 97.00%, and an F1 score of 90.17%. These findings indicate that the Extra Trees algorithm has good performance in analyzing hotel service sentiment based on customer reviews. The implication of this research is to provide guidance to hotels to understand and improve their service quality based on feedback from customers. In addition, this research can also be the basis for further development in the field of sentiment analysis and customer service in the tourism industry.
OPTIMALISASI KINERJA KLASIFIKASI TEKS BERDASARKAN ANALISIS BERBASIS ASPEK DAN MODEL HYBRID DEEP LEARING Salsabila Rabbani; Agustin; Susandri; Rahmiati; M. Khairul Anam
The Indonesian Journal of Computer Science Vol. 13 No. 3 (2024): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i3.4034

Abstract

The conflict between Palestine and Israel has generated strong debates and reactions on social media, including in Indonesia. Public perception of various aspects is certainly important to identify issues in the Palestinian-Israeli conflict. However, the process of manually classifying aspects of the Palestinian-Israeli conflict requires human resources and considerable time. This research aims to explore the views of Indonesians on the Palestinian-Israeli conflict through sentiment analysis based on aspects of Territory, Religion, Politics, and History. Using deep learning technology, specifically a combination model of Convolutional Neural Networks with Long Short-Term Memory (CNN-LSTM), this research analyzes opinion and views data collected from X social media platform (Twitter). This research shows the results of the dataset obtained that the Political aspect dominates more than other aspects. The model evaluation results obtained an accuracy value of 96%, which indicates that the model's ability to classify X users' sentiments towards the Palestinian-Israeli conflict achieved a high level of success.
Klasifikasi Emosi Terhadap Konflik Israel-Palestina Menggunakan Algoritma Gated Recurrent Unit Saputra, Eko Ikhwan; Fatdha, T.Sy. Eiva; Agustin; Junadhi; M. Khairul Anam
The Indonesian Journal of Computer Science Vol. 13 No. 4 (2024): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i4.4106

Abstract

The Israel-Palestine conflict intensified following the October 7, 2023, attack by Hamas on Israel, triggering various emotional reactions on social media. Emotion classification is crucial for understanding public sentiment related to this conflict. This study utilizes 9,917 tweets from platform X (Twitter) to classify emotions such as joy, sadness, anger, fear, disgust, and surprise. The deep learning algorithm used is Gated Recurrent Unit (GRU), developed with three different training and testing data splits: 70:30, 80:20, and 90:10. For text representation, Global Vector (GloVe) word embedding is employed. Given the imbalanced dataset, this study applies the Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance. The research results indicate that the GRU model with a 90:10 data split without using SMOTE achieves the highest accuracy of 75%, followed by the models with 70:30 and 80:20 splits, which each have an accuracy of 73%.
Penerapan Algoritma Convolutional Neural Network Untuk Klasifikasi Penyakit Kanker Kulit Septhya, Dhini; Rahmaddeni; Susanti; Agustin
The Indonesian Journal of Computer Science Vol. 13 No. 4 (2024): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i4.4262

Abstract

The skin is an important organ that protects the human body, so early treatment is essential to prevent diseases such as skin cancer. Skin cancer is a serious disease that can be fatal ana requires high treatment costs. It ranks thirds after cervical cancer and breast cancer in Indonesia, with causative factors including genetics and exposure to UV radiation. Early detection and proper diagnosis are essential to increase the chances of recovery, so skin cancer classification is necessary to avoid delays in treatment. Deep Learning methods, particularly Convolutional Neural Network (CNN), have been shown to provide significant result in image classification with high accuracy. VGG16 and DenseNet121 are two popular CNN architecture used in image classification. This study aims to compare the performance of skin cancer classification using VGG16 and DenseNet121. The result show that the DenseNet121 architecture provides higher accuracy compared to the VGG16 architecture, with 93% accuracy for train data and 79% for test data, while the VGG16 architecture achieves 80% accuracy for train data and 74% for test data. These results show that the DenseNet121 architecture is superior in skin cancer classification, providing important information for more accurate diagnosis