Article Per Year (5 Year)
p-Index From 2021 - 2026
0.835
P-Index
This Author published in this journals
All Journal JURIKOM (Jurnal Riset Komputer)
Ifnu Wisma Dwi Prastya
Universitas Nahdlatul Ulama Sunan Giri, Bojonegoro
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Perbandingan Algoritma Machine Learning untuk Klasifikasi Kopi Menggunakan Data Sensor Electronic Nose dan Tongue Dwi Issadari Hastuti; Mula Agung Barata; Ifnu Wisma Dwi Prastya
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9349

Abstract

Coffee is a leading Indonesian commodity with a diversity of aromas and flavors influenced by variety and region of origin. However, the process of identifying and classifying coffee types is still often carried out conventionally through sensory testing, which is subjective, time-consuming, and dependent on panelist expertise. This situation encourages the need for a more objective and consistent automated approach based on sensor technology and machine learning. This study aims to compare the performance of several machine learning algorithms, namely Logistic Regression, Support Vector Classifier (SVC), and Random Forest, in classifying Indonesian coffee types using multisensor Electronic Nose and Electronic Tongue data. The data used comes from gas, temperature, and pH sensors with a total of 1,503 samples representing ten coffee classes. The preprocessing stage includes data cleaning using the Interquartile Range (IQR) method to remove outliers and noise reduction using the Moving Average method. The results show that the application of data cleaning and noise reduction significantly improves the performance of all classification models. Among the algorithms tested, Random Forest showed the most stable and superior performance in classifying coffee types. These findings confirm that the combination of appropriate data preprocessing and appropriate algorithm selection plays a crucial role in improving the accuracy of machine learning-based coffee classification systems.
Analisis Perbandingan Seleksi Fitur dalam Memprediksi Kelulusan Mahasiswa dengan Menngunakan Artificial Neural Network M. Khoirul Risqi; Ifnu Wisma Dwi Prastya; Mula Agung Barata
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9420

Abstract

Student attrition presents a major challenge in higher education due to its direct impact on academic quality and institutional graduation rates. Detecting students who are likely to withdraw at an early stage is therefore essential to ensure that timely interventions can be made. This study investigates how three distinct feature selection techniques—Chi-Square, Information Gain, and ANOVA—affect the performance of Artificial Neural Networks (ANN) in classifying student outcomes. The data used in the experiment were drawn from academic and administrative records, which had been standardized through Min-Max normalization. The results demonstrate that each method contributes positively, with classification accuracies ranging from 88.71% to 91.37%. Information Gain emerged as the most effective approach, yielding the highest accuracy at 91.37% and a recall score of 97.29%, largely due to its capability to reduce entropy and isolate the most informative variables. ANOVA also performed consistently well with 90.82% accuracy, while Chi-Square was comparatively less effective, potentially due to its reliance on categorical variables that may not capture predictive nuances. These findings emphasize the strategic importance of applying robust feature selection to improve ANN-based prediction models. Ultimately, this research supports the design of data-driven systems aimed at reducing student dropout rates and strengthening academic retention strategies across higher education institutions.
Analisis Sentimen Multi-Platform Media Sosial pada Program Makan Bergizi Gratis Menggunakan Ensemble IndoBERT-SVM Muhammad Bisri Mustofa; Ifnu Wisma Dwi Prastya; Sahri Sahri
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9460

Abstract

Sentiment analysis of public policy on social media faces significant challenges due to linguistic heterogeneity across platforms and limitations of single models in capturing the diversity of opinion expressions. Previous studies tend to employ single-platform and single-model approaches that potentially generate representational bias and accuracy degradation of up to 12–15% when applied to different platform contexts. This study aims to develop a soft voting-based ensemble model that integrates Support Vector Machine (SVM) and IndoBERT to analyze public sentiment toward the Free Nutritious Meal (MBG) Program across multiple platforms, and to evaluate the effectiveness of the ensemble approach compared to single models in addressing variations in linguistic characteristics of digital platforms. The research dataset consists of 7,500 comments from X, TikTok, and YouTube collected from January 6 to September 28, 2025, processed through informal Indonesian language preprocessing, lexicon-based labeling, and stratified split division. Results demonstrate that SVM performs optimally on TikTok (accuracy 98.1%, macro F1 98.0%) but weakly on the neutral class in X (F1 51.0%), while IndoBERT excels in handling pragmatic ambiguity in X (neutral F1 74.0%) despite slightly declining on TikTok (macro F1 93.0%). The ensemble model produces the most balanced performance with accuracies of 92.53% (YouTube), 95.73% (TikTok), 92.53% (X), and macro F1 scores of 85.07%, 94.33%, 84.92% respectively. The contributions of this research include the development of a multi-platform sentiment analysis approach that addresses single-platform bias, improved classification generalization capability across heterogeneous digital ecosystems, and provision of evidence-based evaluation instruments for improving government policy implementation and communication.
Analisis Sentimen Komentar Cyberbullying Terhadap Fenomena Flexing di Tiktok Menggunakan Artificial Neural Network Lailatul Qodriyah; Ifnu Wisma Dwi Prastya; Guruh Purbo Dirgontoro
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9494

Abstract

The rising trend of flexing on TikTok has created a dynamic digital space that often triggers varied user reactions, including subtle forms of cyberbullying. This study aims to analyze public sentiment toward flexing content and evaluate the performance of the Artificial Neural Network (ANN) algorithm in classifying user comments. A total of 4,013 comments were collected through a scraping process on the TikTok account of Miechel Halim and automatically labeled using a lexicon-based approach. The comments were then pre-processed and transformed into Term Frequency–Inverse Document Frequency (TF-IDF) representations before being split into training and testing datasets with an 80:20 ratio. The ANN model was trained under two scenarios before and after the application of the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance. Experimental results show that the initial model achieved an accuracy of 89.79%, which increased to 90.04% after SMOTE, accompanied by an improvement in the recall of the negative class. These findings indicate that ANN is effective for sentiment classification of TikTok comments, although informal language patterns and highly imbalanced labels remain challenges in identifying negative or potentially harmful remarks related to cyberbullying.
Perbandingan Metode Euclidean dan Manhattan Distance dalam Implementasi Algoritma K-Means dan K-Medoid pada Pengelompokkan Faktor Dominan Perceraian di Kabupaten Bojonegoro Elok Salma Nabila Salma; Ifnu Wisma Dwi Prastya; Ita Aristia Sa’ida
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9520

Abstract

The divorce rate in Bojonegoro Regency continues to increase, driven by various social factors such as constant disputes, economic pressure, and household disharmony. Consequently, an analysis is required to map dominant and non-dominant factors more effectively. This study aims to group the factors causing divorce in Bojonegoro Regency for the 2021–2023 period and determine the most optimal clustering method. The research utilizes K-Means and K-Medoids algorithms with Euclidean and Manhattan distance metrics applied to both raw data and data normalized using the Min–Max Scaler, evaluated via the Silhouette Score. The results indicate that data normalization improves cluster quality, and K-Means with Manhattan distance on normalized data achieves the best performance, yielding a Silhouette Score of 0.849547. Cluster displacement analysis reveals that the grouping patterns remain relatively consistent across years, with "constant disputes" consistently emerging as the dominant factor, while other factors remain in the non-dominant cluster with similar patterns. This study demonstrates that K-Means with Manhattan distance on normalized data is more effective for clustering divorce factors. These findings can serve as a methodological foundation for the local government in formulating data-driven social policies and interventions.
Co-Authors Dwi Issadari Hastuti Elok Salma Nabila Salma Guruh Purbo Dirgontoro Ita Aristia Sa’ida Lailatul Qodriyah M. Khoirul Risqi Muhammad Bisri Mustofa Mula Agung Barata Sahri Sahri