cover
Contact Name
Mustakim
Contact Email
Mustakim
Phone
-
Journal Mail Official
ijaidm@uin-suska.ac.id
Editorial Address
-
Location
Kab. kampar,
Riau
INDONESIA
Indonesian Journal of Artificial Intelligence and Data Mining
ISSN : 26143372     EISSN : 26146150     DOI : -
Core Subject : Science,
Indonesian Journal of Artificial Intelligence and Data Mining (IJAIDM) is an electronic periodical publication published by Puzzle Research Data Technology (Predatech) Faculty of Science and Technology UIN Sultan Syarif Kasim Riau, Indonesia. IJAIDM provides online media to publish scientific articles from research in the field of Artificial Intelligence and Data Mining. IJAIDM will be published 2 (two) times a year, in March and September, each edition contains 7 (seven) articles. Articles may be written in English or Indonesia.
Arjuna Subject : -
Articles 233 Documents
Identifying Twitter Topics Using K-Means Clustering and Association Rule Mining for Improved Insights Lengari, Cristiany Gunu; Puspitasari, Ira
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 1 (2025): March 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i1.31720

Abstract

The annual growth in social media users has led businesses to increasingly leverage these platforms for marketing, promotion, and addressing public complaints. Twitter, now known as X, stands out as one of the most widely used social media platforms. It serves as a forum for various opinions and complaints regarding services provided by businesses. This study focuses on analyzing public opinions related to Indihome services, as expressed on the @indihomecare Twitter account. These opinions range from expressions of support to complaints about internet services and Indihome's responses to these issues. This study employs a text clustering approach using the K-means algorithm on Twitter data, complemented by association rules to identify topics related to Indihome customer complaints. The optimal number of clusters is determined using the Elbow method, while Word Cloud visualizations are utilized to illustrate frequently occurring words within each cluster. The application of association rules revealed that the most frequently appearing words, with a support value of 0.057, were "indihome," "account," "whatsapp," and "channel." These findings provide insights into the primary concerns and communication channels used by Indihome customers on Twitter
Analysis of The Level of Satisfaction of Electric Bus Passengers in Medan Using C5.0 Algorithm Ritonga, Azizah Oktarina; Sriani, Sriani
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.32785

Abstract

The C5.0 method was successfully implemented in the analysis of passenger satisfaction level of Medan City Electric Bus using 500 passenger data divided into 70% for training data (350 data) and 30% for test data (150 data). The steps include creating a decision tree based on training data, where the model learns to identify patterns that affect passenger satisfaction. After the decision tree is formed, the model is tested using testing data to measure the prediction accuracy. The evaluation results show that the C5.0 model is able to classify the testing data effectively, providing an accurate picture of passenger satisfaction levels in the tested context. Based on manual calculations with 10 testing data against training data, the following prediction results are obtained: out of 10 testing data, the model predicts 7 data as Satisfied (satisfied) and 3 data as Dissatisfied (not satisfied). This result shows that the model managed to classify most of the testing data correctly, giving a positive indication of the accuracy and reliability of the model in identifying passenger satisfaction levels. Based on the evaluation results of the C5.0 model using R, the training data showed accuracy, precision, and recall each reached 100%. This indicates that the C5.0 model was perfectly successful in classifying all training data, with no errors in prediction or classification. This result confirms that the model is very effective and reliable in analyzing the satisfaction level of Medan City Electric Bus passengers, demonstrating its ability to provide accurate and consistent predictions.
Analysis of Social Vulnerability in Java Island using K-Medoids Algorithm with Variation of Distance Measurements (Euclidean, Manhattan, Minkowski) Nur, Indah Manfaati; Abdurakhman, Abdurakhman
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.31111

Abstract

The Social Vulnerability Index (SoVI) measurement to assess social vulnerability is only able to describe conditions in general, without being able to show which factors dominate the score. Therefore, the aim of this research is to fill this gap by applying a correlational approach with a clustering method to characterize the dominant factors of social vulnerability at the district level in Java and surrounding areas. The clustering method used in this study is the K-Medoids algorithm. This method is more powerful when there are outliers in the dataset used. In this study, we considered the use of 3 different distance methods, namely Euclidean distance, Manhattan distance, and Minkowski distance. As a result, the K-Medoids algorithm using Manhattan distance provides the best value based on the Davies Bouldin Index. This research found that social vulnerability exists in every region of Java Island and its surroundings.
Design and Development of An Intelligent Automatic Tilapia Fish Farming Device in A Bucket Based on Internet of Things Suni, Gina Amanah; Fadhli, Mohammad; Rose, Martinus Mujur
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 1 (2025): March 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i1.31809

Abstract

The cultivation of various freshwater fish species, such as catfish, tilapia, carp, and sepat, can be effectively managed through the budikdamber technique, where fish and vegetables are grown together in a single container. This research introduces an alternative method designed to control water temperature, automate fish feeding, and cover the container automatically when it rains. By integrating monitoring and control devices, budikdamber owners can manage automated feeding, monitor water temperature, measure pH levels, control water depth, and automatically activate rain covers. This smart device is expected to enhance budikdamber management efficiency, contributing to the improved welfare of the fish and overall system sustainability.
Early Detection of Phishing Sites with Enhanced Neural Network Models Suarti, Isa; Chamidy, Totok; Crysdian, Cahyo
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.30068

Abstract

Phishing is a digital crime committed with the aim of obtaining personal data by creating a link or website that resembles the original. This form of cyber attack is caused by a notification in a text message, email, or phone call. A common anti-phishing countermeasure technique is to perform early detection of potentially phishing sites, primarily according to the source code features, which are required to traverse web page content, as well as third parties that slow down the process of clarifying phishing URLs. Although the latest technology has long been used in phishing early detection, there is still a need for manual feature engineering that is important and reliable enough to detect emerging phishing offenses. One of these involves training a neural network (NN) using a dataset of known phishing URLs and legitimate URLs. The research was conducted using 200 data, Data were separated into training and testing categories.  Training was done using 100 and 120 data. Training results on 100 data and 160 data had lower iterations and errors on the tanh activation function compared to the logistic activation function. The number of iterations that occur in logistic activation is as many as 400 iterations, while when using the tanh activation function only 175 iterations are needed.
Single Moving Average Method Forecasting to Predict Skincare Sales Marhama, Siti; Sembiring, Muhammad Ardiansyah; Sena, Maulana Dwi
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 1 (2025): March 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i1.32028

Abstract

Nada Nadine Skincare Store is a business engaged in the sale of beauty skincare. Nada Nadine Skincare Store was established in 2023 and is located on Jalan Lingkungan IV Air Joman, Asahan Regency. The problem at Nada Nadine Skincare Store is that it has difficulty in estimating the level of demand for beauty skincare (Wardah) for inventory because it is still guessing and experience from the production department of Nada Nadine Skincare Store and there is no beauty skincare sales forecasting system that will make it easier for Nada Nadine Skincare Store to determine the amount of beauty skincare (Wardah) inventory. The purpose of this system is to be able to apply the single moving average method to the beauty skincare (Wardah) sales demand forecasting system at the Nada Nadine Skincare Store and to be able to design a website-based beauty skincare (Wardah) sales demand forecasting system at the Nada Nadine Skincare Store using the PHP programming language and MySql database. From the problem analysis, the author tries to predict the amount of beauty skincare demand using the Single Moving Average (SMA) method. The conclusion that can be drawn from research on the Single Moving Average (SMA) based skincare sales forecasting system forecasts the sales demand for beauty skincare (wardah) at the Nada Nadine Skincare Store so The system calculation results are the same as the manual calculation forecast July 2024 is 33,50, forecast August 2024 is 32,75, MAD value is 2,70, MSE value is 11,31 and MAPE value is 8,84%.
Forecasting Climate Change Patterns to Improving Rice Harvest Using SVR for Achieving Green Economy Juliandy, Carles; Kelvin, Kelvin; Halim, Apriyanto; Pipin, Sio Jurnalis; Sinaga, Frans Mikael; Lestari, Wulan Sri
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.32393

Abstract

The consistently declining rice harvest will cause several economic and environmental problems. The unstable and unpredictable climate change was believed as the main problem of the declining rice harvest. We proposed a method for forecasting climate change to help the farmer in their rice cultivation. We used Support Vector Regression (SVR) to improve algorithm steps such as normalizing the data and applying an Adaptive Linear Combiner (ALC) to optimize the dataset before we processed it with the algorithm. Our model gets 95% accuracy as measured with the confusion matrix. We believe our model will help the farmers in their rice cultivation with good climate forecasting. A further benefit of this research we belief that with the well-forecasted climate, the usage of pesticides will decrease and will help the vision of the Indonesian government with a green economy
Machine Learning Approach for Early Diagnosis of Dyslexia Among Primary School Children: A Scoping Review and Model Development Kurniawan, Zaqi; Tiaharyadini, Rizka
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.30614

Abstract

Dyslexia, a prevalent learning disorder among primary school children, often goes undetected until later stages, hindering academic progress and socio-emotional development. Early diagnosis is crucial for effective intervention. Machine Learning (ML) offers promise in developing accurate diagnostic tools. However, there's a scarcity of comprehensive reviews focusing on ML approaches for dyslexia diagnosis in this demographic. In this scoping review, we consolidate existing literature and present the development of a novel ML model that was customized for early dyslexia diagnosis. Utilizing Decision Tree, K-Nearest Neighbors (KNN), Logistic Regression, Naive Bayes, and Random Forest. The comparative analysis of ML methods for dyslexia detection in elementary school children reveals distinct strengths. Decision Tree shows robust precision: 92.31% for dyslexia-prone, 90.62% for diagnosed dyslexia, and 86.67% for no dyslexia detected, with corresponding high recall values of 90.57%, 87.88%, and 100%, respectively. KNN excels with an overall accuracy of 94.00% and perfect precision for undetected dyslexia (100%), with high precision and recall for dyslexia-prone and diagnosed dyslexia. Logistic Regression highlights significant predictors and achieves precision of 95.38% for dyslexia-prone and 88.24% for diagnosed dyslexia, with recall rates of 93.34% and 90.91%, respectively. Naive Bayes exhibits outstanding precision for no dyslexia and dyslexia-prone categories (100%), with slightly lower precision for diagnosed dyslexia (82.5%), but perfect recall for undetected and diagnosed dyslexia. Random Forest demonstrates balanced performance with precision ranging from 91.18% to 94.23% and recall from 92.31% to 93.94%, achieving an overall accuracy of 93.00%. These results underscore ML's potential in enabling early dyslexia detection, facilitating timely interventions to improve outcomes for affected children and advancing dyslexia diagnosis.
Performance Evaluation of Machine Learning Models for Predicting Household Energy Consumption: A Comparative Study Airlangga, Gregorius
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 1 (2025): March 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i1.32791

Abstract

Accurate prediction of household energy consumption is critical for improving energy efficiency and optimizing resource allocation in smart grids. This study evaluates the performance of several machine learning regression models, including Linear Regression, Ridge Regression, Lasso Regression, Random Forest, Gradient Boosting, XGBoost, CatBoost, and LightGBM, for predicting daily household energy consumption. The models were trained and tested on time series data, and their performance was measured using four key metrics: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R². Results show that non-linear models, especially ensemble-based methods such as Random Forest and CatBoost, outperformed traditional linear regression models. Random Forest achieved the lowest MAE (0.1682) and competitive RMSE (0.2450), making it the best overall model. CatBoost, with its advanced gradient boosting algorithm, also demonstrated superior predictive accuracy, achieving an RMSE of 0.2421 and an MAE of 0.1830. In contrast, linear models struggled to capture the complex patterns in the data, with Linear Regression showing the worst performance. The negative R² scores across all models indicate challenges in explaining the variance in the dataset, which may be attributed to external factors or noise not captured by the models. This study highlights the importance of choosing appropriate machine learning models for time series forecasting and recommends further exploration of deep learning models and external features to improve prediction accuracy.
Leveraging Machine Learning for Accurate Anemia Diagnosis Using Complete Blood Count Data Airlangga, Gregorius
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.29869

Abstract

Anemia, a prevalent hematologic disorder, necessitates accurate and timely diagnosis for effective management and treatment. This study explores the application of various machine learning models to classify anemia types using complete blood count (CBC) data. We evaluated multiple models, including DecisionTreeClassifier, ExtraTreeClassifier, RandomForestClassifier, ExtraTreesClassifier, XGBoost, LightGBM, and CatBoost, to identify the most effective approach for anemia diagnosis. The dataset comprised CBC data labeled with anemia diagnoses, sourced from multiple medical facilities. Rigorous data preprocessing was performed, followed by feature selection using methods such as Variance Inflation Factor (VIF), Predictive Power Score (PPS), and feature importance from ensemble models. The models were trained and evaluated using 5-fold cross-validation, with hyperparameter tuning conducted via GridSearchCV. Results demonstrated that the DecisionTreeClassifier achieved the highest balanced accuracy score of 94.17%, outperforming more complex ensemble methods. Confusion matrices validated its robust performance, highlighting its precision and recall. The study underscores the potential of simple decision tree models in medical diagnosis tasks, particularly when datasets are well-preprocessed. These findings have significant implications for clinical practice, suggesting that machine learning can enhance diagnostic accuracy and efficiency. Future work will explore advanced techniques to further improve performance and integration into clinical workflows.