Jasmir , Jasmir
Unknown Affiliation

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

Sistem Pendukung Keputusan Keluarga Penerima Manfaat Bantuan Langsung Tunai Dana Desa (BLT-DD) Kusuma Safitri, Lestari; Jasmir , Jasmir
Jurnal Manajemen Sistem Informasi Vol 8 No 4 (2023): MANAJEMEN SISTEM INFORMASI
Publisher : LPPM Universitas Dinamika Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33998/jurnalmsi.2023.8.4.1519

Abstract

Village Fund Direct Cash Assistance is the government's effort to stabilize the economy of the poor in Indonesia, especially in villages. collectively and as social protection for the community. The purpose of this study was to design and apply a decision support system to help determine a decision for the Village Head of Bukit Tempurung to determine the prospective beneficiary families of the Village Fund Direct Cash Assistance (BLT-DD). This study uses a research method using SAW (Simple Addictive Weight). This research will produce a ranking of each alternative. Based on the research that has been done, there are still difficulties because the assessment is still done subjectively. This study uses UML (Unified Modeling Language) tools in the form of use case diagrams and activity diagrams. With this decision support system it can help minimize errors, because the determination is based on predetermined criteria and has gone through a calculation process so that the assessment is objective
Comparison of ANOVA and Chi-Square Feature Selection Methods to Improve Machine Learning Performance in Anemia Classification Annisa, Tiko Nur; Jasmir , Jasmir; Nurhadi , Nurhadi
Jurnal Teknik Informatika (Jutif) Vol. 6 No. 4 (2025): JUTIF Volume 6, Number 4, Agustus 2025
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2025.6.4.5017

Abstract

Anemia is a prevalent hematological condition marked by decreased hemoglobin concentration in the blood, which can lead to serious health complications if undetected. Although machine learning has shown potential in supporting early diagnosis, its effectiveness is often hindered by irrelevant or excessive features. This study investigates the impact of ANOVA and Chi-Square feature selection methods in improving the effectiveness of three distinct machine learning models algorithms, Naive Bayes, K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) for anemia classification. Using a Kaggle dataset consisting of 15,300 instances and 25 features, the evaluation of each model was conducted with reference to its accuracy, precision, recall, and F1-score, both before and after applying feature selection. Experimental results show a substantial improvement in classification performance after feature selection, with the SVM + ANOVA combination achieving the highest accuracy of 94.61%. In contrast, models without feature selection performed below 90%, highlighting the need for appropriate feature reduction techniques. This study contributes a comparative analysis framework for medical data classification, emphasizing the role of statistical feature selection in optimizing model accuracy. Its novelty lies in demonstrating consistent performance improvement across algorithms using real-world anemia data and providing evidence that ANOVA and Chi-Square can significantly enhance model generalization in medical diagnostic contexts.
Fitur Information Gain untuk Meningkatkan Nilai Performa Pengklasifikasi Machine Learning pada Analisis Sentimen Komentar Spam Pengguna Youtube Jasmir, Jasmir; Gunardi, Gunardi; Rohaini, Eni; Naibaho, Ronald; Sukoco, Bambang; Jasmir , Jasmir
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 13 No 2: April 2026
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.132

Abstract

Perkembangan pesat media sosial telah memberikan ruang bagi setiap individu untuk menyampaikan pendapat, baik berupa komentar positif maupun negatif terhadap konten yang mereka akses. Kemudahan dalam memberikan opini secara daring ini berdampak pada semakin besarnya jumlah ulasan yang tersedia. Namun, volume ulasan yang sangat besar sering kali sulit untuk dianalisis secara manual dan berpotensi menimbulkan bias dalam penilaian. Untuk mengatasi permasalahan tersebut, diperlukan pendekatan otomatis melalui klasifikasi sentimen yang bertujuan mengelompokkan opini pengguna ke dalam kategori positif atau negatif. Dalam penelitian ini digunakan tiga algoritma pembelajaran mesin, yaitu Naïve Bayes (NB), K-Nearest Neighbor (KNN), dan Random Forest (RF). Data penelitian diperoleh dari public dataset UCI Machine Learning. Fokus penelitian adalah meningkatkan kinerja klasifikasi dengan memanfaatkan teknik seleksi fitur information gain. Hasil eksperimen menunjukkan bahwa penerapan information gain secara konsisten meningkatkan performa semua algoritma yang diuji, baik pada metrik akurasi, presisi, recall, maupun f1-score. Naïve Bayes awalnya memperoleh akurasi tertinggi sebesar 74,33% pada kondisi tanpa fitur tambahan. Namun, setelah penerapan information gain, algoritma KNN menunjukkan hasil paling optimal dengan akurasi mencapai 81,28% serta performa yang relatif seimbang pada semua metrik evaluasi. Sementara itu, Random Forest juga mengalami peningkatan, meskipun tidak melampaui KNN. Secara keseluruhan, penelitian ini menegaskan bahwa pemilihan fitur yang relevan melalui information gain mampu meningkatkan efisiensi dan efektivitas klasifikasi sentimen, serta dapat menjadi pendekatan yang potensial untuk menganalisis opini dalam skala besar.   Abstract The rapid growth of social media has provided individuals with the opportunity to freely express their opinions, whether positive or negative, toward the content they encounter. The increasing ease of sharing opinions online has resulted in a massive volume of user reviews. However, the large number of reviews is difficult to analyze manually and may introduce bias in interpretation. To address this issue, sentiment classification is applied to automatically categorize user opinions into positive or negative classes. In this study, three machine learning algorithms were employed: Naïve Bayes (NB), K-Nearest Neighbor (KNN), and Random Forest (RF). The dataset was obtained from the public UCI Machine Learning repository. The main objective of this research is to improve classification performance by utilizing feature selection through the information gain method. Experimental results demonstrate that applying information gain consistently enhances the performance of all evaluated algorithms across multiple metrics, including accuracy, precision, recall, and F1-score. Without feature selection, Naïve Bayes achieved the highest accuracy of 74.33%. However, after applying information gain, KNN outperformed the other algorithms by reaching an accuracy of 81.28% and exhibited balanced results across all evaluation metrics. Random Forest also showed improvement but did not surpass the performance of KNN. Overall, these findings highlight the importance of feature selection in improving both the efficiency and effectiveness of sentiment classification. Furthermore, the use of information gain proves to be a promising approach for large-scale opinion analysis, particularly in handling the high dimensionality of textual data.