Claim Missing Document
Check
Articles

Applying TF-IDF and K-NN for Clickbait Detection in Indonesian Online News Headlines Afif, Muhammad Athallah; Ula, Munirul; Rosnita, Lidya; Rizal, Rizal
Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024
Publisher : Department of Informatics, Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/jacka.v1i2.15810

Abstract

This research explores the application of TF-IDF (Term Frequency-Inverse Document Frequency) and K-Nearest Neighbor (K-NN) in constructing a clickbait detection system for Indonesian online news headlines. The TF-IDF method is employed to ascertain the significance of words in news headlines, utilizing a tokenization process to generate numeric representations. The TF-IDF matrix serves as features in the K-NN classification model, with k=1 determining the most similar class. Model evaluation yields outstanding results, achieving accuracy, precision, recall, and F1-Score all reaching 1.0. The confusion matrix unveils no misclassifications, affirming the model's adeptness in correctly classifying all samples.
Application of Fuzzy C-Means and Borda in Clustering Crime–Prone Areas and Predicting Crime Rates Using Long Short Term Memory in Northern Aceh Regency Lubis, Syahrul Andika; Ula, Munirul; Retno, Sujacka
International Journal of Engineering, Science and Information Technology Vol 5, No 1 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i1.747

Abstract

North Aceh is a district with diverse geographical conditions, ranging from vast lowland areas in the north stretching from west to east, to mountainous areas in the south. The average altitude in North Aceh is 125 meters. The district covers an area of 2,694.66 km² with a population of 614,640 people in 2022. The issue of crime in North Aceh District has caused significant discomfort among the community. According to data from the Central Bureau of Statistics (BPS) of Aceh Province, the number of criminal cases increased from 6,651 cases in 2022 to 10,137 cases in 2023. Using the Fuzzy C-Means clustering method, the data was grouped into three clusters: cluster 1 represents safe areas, cluster 2 represents moderately vulnerable areas, and cluster 3 represents vulnerable areas. For ranking using the Borda method, the Dewantara Police Sector ranked first for the physical aspect, while the Muara Batu Police Sector ranked first for the item aspect. As for predictions using the LSTM model, almost all subdistricts achieved MAPE values below 20%, indicating that the LSTM model is quite effective in predicting crime-prone areas. For example, Baktiya District recorded a MAPE value of 15.85% for the physical aspect, while the best result was achieved by Simpang Keramat District for the item aspect with a MAPE value of 0.00%. However, in Syamtalira Bayu District, the item aspect reached a MAPE value of 20.07%. Although the MAPE value for the item aspect in Syamtalira Bayu is relatively high, it is still considered acceptable as it remains below 50%.
ANALISIS KINERJA TATA KELOLA TEKNOLOGI INFORMASI MENGGUNAKAN FRAMEWORK COBIT 2019 PADA UNIVERSITAS JABAL GHAFUR Salimuddin, Salimuddin; Ula, Munirul; Nurdin, Nurdin
Jurnal Informatika dan Teknik Elektro Terapan Vol 13, No 2 (2025)
Publisher : Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jitet.v13i2.6130

Abstract

This research aims to evaluate Information Technology Governance at Jabal Ghafur University (Unigha) using the COBIT 2019 Framework. The focus of the research includes analysis of Information Technology operational processes, measurement of feasibility with the COBIT 2019 Design Factor Toolkit, and performance evaluation on two main process objectives, namely EDM03 (Ensured Risk Optimization) and MEA03 (Managed Compliance with External Requirements). This research involved respondents selected based on RACI Chart analysis, consisting of the Vice Chancellor I, Head of the General Administration Bureau, Head of the Administration Section, Head of PUKSI, Head of the Information Security Section, and Head of the Quality Assurance Agency (LPM) using a questionnaire. The analysis results show that these two process objectives have an average capability value of 100% at Capability Level 1, but only achieved Largely Achieved at Capability Level 2. The gap analysis shows a gap between the current condition (Level 1) and the desired target (Level 4), with a difference of 3. Based on these findings, it is recommended that Unigha strengthen risk management and compliance with external requirements, through updating internal policies, improving HR training and utilizing technology more effectively. This improvement is expected to increase the level of capability and performance of Information Technology Governance in Unigha.
Internet of Things and Artificial Neural Network Application for Optimizing Spirulina Cultivation with Palm Oil Mill Effluent Ula, Munirul; Fajriana, Fajriana; Ulfah, Julia
Journal of Computer Science, Information Technology and Telecommunication Engineering Vol 6, No 1 (2025)
Publisher : Universitas Muhammadiyah Sumatera Utara, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30596/jcositte.v6i1.22389

Abstract

This study aims to optimize algae biomass production by utilizing Palm Oil Mill Effluent (POME) as a nutrient source, employing Internet of Things (IoT) technology and Artificial Neural Networks (ANN) for predictive modeling and system control. POME, an organic waste from the palm oil industry, was used as an organic liquid fertilizer to enhance the efficiency and sustainability of algae cultivation. The system was designed to monitor and control key environmental parameters such as pH, temperature, salinity, and dissolved oxygen in real-time during a one-month trial in July 2024. ANN-based models were used to predict and adjust environmental conditions, leading to significant improvements in algae growth and resource efficiency. The results indicate that POME can serve as an effective and eco-friendly nutrient source, contributing to both reduced industrial waste and sustainable biomass production. This integrated approach supports circular economy principles and sustainability goals, with potential applications in bioresource production and waste management. Future research will focus on large-scale system testing, optimization for various algae species, and long-term sustainability assessment.
Comparing Long Short-Term Memory and Random Forest Accuracy for Bitcoin Price Forecasting Munirul Ula; Veri Ilhadi; Zailani Mohamed Sidek
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 23 No. 2 (2024)
Publisher : LPPM Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v23i2.3267

Abstract

Bitcoin’s daily value fluctuations are very dynamic. Understanding its rapid and intricate price movements demands advanced techniques for processing complex data. This research aims to compare the accuracy of two machine learning methods, Random Forest (RF) and Long Short-Term Memory (LSTM), in predicting Bitcoin price. This research employs RF and LSTM algorithms to forecast Bitcoin prices using a two-year Yahoo Finance dataset. The evaluation metrics used were accuracy based on Mean Absolute Percentage Error (MAPE) and computational power (CPU-Z). As a result of this research, the LSTM model demonstrates higher accuracy compared to the RF model. MAPE reveals LSTM’s precision of 99.8% and RF’s accuracy of 90.1%. Regarding computational time and resources, RF shows slightly better performance than LSTM. The visual comparison further emphasizes LSTM’s better performance in predicting Bitcoin prices, highlighting its potential for informed decision-making in cryptocurrency trading. This research contributes valuable insights into the effectiveness, strengths, and weaknesses of LSTM and RF models in predicting cryptocurrency trends.
Perbandingan Kinerja Protokol MQTT dan HTTP Dalam Komunikasi Data Internet of Things Fikhri, Aditya Aziz; Ula, Munirul; Sayuti, Muhammad; Taufiq, Taufiq; Nudin, Nurdin
Jurnal Infomedia: Teknik Informatika, Multimedia, dan Jaringan Vol 10, No 1 (2025): Jurnal Infomedia
Publisher : Politeknik Negeri Lhokseumawe

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30811/jim.v10i1.6733

Abstract

Penelitian ini membandingkan kinerja protokol MQTT dan HTTP dalam sistem komunikasi Internet of Things (IoT), khususnya untuk pemantauan kualitas udara ruang kelas secara real-time. Evaluasi dilakukan menggunakan server virtual machine dengan spesifikasi identik, berdasarkan parameter seperti penggunaan CPU, waktu pengiriman pesan, dan tingkat kehilangan data. MQTT, sebagai protokol ringan dengan model publish-subscribe, menunjukkan kecepatan pengiriman pesan yang jauh lebih tinggi dibandingkan HTTP, terutama pada skenario dengan volume pesan yang besar. Namun, penggunaan CPU pada MQTT meningkat tajam seiring bertambahnya jumlah pesan, dan terjadi kehilangan data yang signifikan hingga 33,8% pada pengiriman 600.000 pesan. Sebaliknya, HTTP yang berbasis model request-response dengan mekanisme multi-proses, mampu menjaga keandalan pengiriman pesan hingga 100%, meskipun waktu pengirimannya jauh lebih lambat. Hasil penelitian ini menunjukkan bahwa MQTT lebih efisien untuk sistem yang membutuhkan kecepatan tinggi dan dapat mentoleransi sebagian kehilangan data, sementara HTTP lebih cocok untuk aplikasi yang menuntut keandalan tinggi dan akurasi data secara penuh. Temuan ini memberikan wawasan penting bagi pengembang dalam memilih protokol komunikasi yang sesuai berdasarkan kebutuhan sistem IoT dan skala implementasinya.
ANALISA DAN DETEKSI KONTEN HOAX PADA MEDIA BERITA INDONESIA MENGGUNAKAN MACHINE LEARNING Ula, Munirul
Jurnal Teknologi Terapan and Sains 4.0 Vol 1 No 2 (2020): Jurnal Teknologi Terapan & Sains
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/tts.v1i2.3263

Abstract

Sekarang  ini konten Hoax yang mengandung informasi tidak benar malah sering kali menjadi konsumsi massal pengguna internet. Hal ini merupakan sesuatu yang buruk karena dapat meningkatkan rasa tidak percaya terhadap berita dan informasi yang ada di internet hingga menimbulkan kebingungan pada masyarakat dalam menentukan informasi mana yang benar. Dalam Penelitian ini, percobaan yang dilakukan bertujuan untuk memilih algoritma terbaik dalam membedakan berita hoax dan berita asli menggunakan metode text mining serta pendekatan dengan machine learning dan  150 artikel berbahasa Indonesia (50 artikel hoax dan 100 artikel asli) sebagai data yang akan digunakan.Penelitian ini akan dimulai dengan tahap preprocessing teks yang terdiri dari tokenizing, case folding, filtering, stopword removal, stemming dan weighting TF-IDF menggunakan penggabungan fitur unigram dan bigram baru kemudian diolah menjadi teks klasifikasi. Hasil dari penelitian ini didapatkan kesimpulan bahwa  algoritma Random Forest memiliki akurasi terbaik dalam mengklasifikasikan berita hoax dan berita asli dibandingkan dengan algoritma Multilayer Perceptron, Naïve Bayes,dan Support Vector Machine dengan nilai akurasi 75.37%. Kata kunci : Klasifikasi,  Berita, Hoax,  Text mining,  Machine learning
Image Feature Extraction for Determining the Ripeness Level OF Oil Palm Fruits Using the K-Nearest Neighbor Algorithm Based on Color Features (Case PTPN IV Aceh Utara) Sudarti, Atrida; Ula, Munirul; Fajriana, F
IJISTECH (International Journal of Information System and Technology) Vol 9, No 1 (2025): The June Edition
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v9i1.391

Abstract

The availability of oil palm fruits at the appropriate ripeness level is crucial to achieving optimal oil production. Farmers often struggle to accurately determine fruit ripeness, resulting in inconsistent quality and reduced efficiency. This study aims to develop a classification system to determine the ripeness level oil palm fruits using the K-Nerest Neighbor (K-NN) algorithm based on color features extracted from fruits image. Color is a key indicator of maturity and directly influences oil yield. The data was collected through image acquisition and direct observation at the Cot Girek Palm Oil Mill (PKS) of PTPN IV, Aceh Utara. Image preprocessing was carried out to enhance and nomalize the data before feature extraction. The extracted color features were then used to classify the fruits into ripe and unripe categories using the K-NN algorithm. The results show that K-NN successfully classifies the ripeness level of oil palm fruits with an accuracy of 72.80%. This system provides a recommendation for fruit feasibility before processing, helping reduce production losses caused by immature or overripe fruits. Overall, this research contributes to improving decision-making in the palm oil industry through the application of image processing of machine learning techniques.
A Comparative Study of Temporal Convolutional Network and Gated Recurrent Unit for Predicting Ethereum Prices Kiram, Saiful; Ula, Munirul; Kurniawati, Kurniawati
Applied Engineering, Innovation, and Technology Vol. 2 No. 1 (2025)
Publisher : MSD Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62777/aeit.v2i1.55

Abstract

This study compares the performance of the Temporal Convolutional Network (TCN) and Gated Recurrent Unit (GRU) models in predicting the price of Ethereum, which is important to support cryptocurrency investment strategies. With the high volatility of the cryptocurrency market, an accurate and reliable prediction model is needed. In this study, Ethereum's daily closing price data over four years was analyzed using TCN and GRU models to evaluate its predictive capabilities. Model accuracy is measured using Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Mean Squared Error (MSE). The results showed that the TCN model excelled in average accuracy with lower MAE and MAPE values, while the GRU model showed excellence in reducing the impact of large errors with smaller MSE values. This reflects TCN's superiority in capturing the overall pattern of price movements, while the GRU is more responsive to short-term price fluctuations. These findings demonstrate the potential of both models in cryptocurrency price forecasting, with their respective advantages. This research provides valuable information for investors and researchers in developing predictive strategies in dynamic financial markets. A combination of TCN and GRU models can also be explored to improve prediction performance in the future.
Image Feature Extraction for Determining the Ripeness Level OF Oil Palm Fruits Using the K-Nearest Neighbor Algorithm Based on Color Features (Case PTPN IV Aceh Utara) Sudarti, Atrida; Ula, Munirul; Fajriana, F
IJISTECH (International Journal of Information System and Technology) Vol 9, No 1 (2025): The June Edition
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v9i1.391

Abstract

The availability of oil palm fruits at the appropriate ripeness level is crucial to achieving optimal oil production. Farmers often struggle to accurately determine fruit ripeness, resulting in inconsistent quality and reduced efficiency. This study aims to develop a classification system to determine the ripeness level oil palm fruits using the K-Nerest Neighbor (K-NN) algorithm based on color features extracted from fruits image. Color is a key indicator of maturity and directly influences oil yield. The data was collected through image acquisition and direct observation at the Cot Girek Palm Oil Mill (PKS) of PTPN IV, Aceh Utara. Image preprocessing was carried out to enhance and nomalize the data before feature extraction. The extracted color features were then used to classify the fruits into ripe and unripe categories using the K-NN algorithm. The results show that K-NN successfully classifies the ripeness level of oil palm fruits with an accuracy of 72.80%. This system provides a recommendation for fruit feasibility before processing, helping reduce production losses caused by immature or overripe fruits. Overall, this research contributes to improving decision-making in the palm oil industry through the application of image processing of machine learning techniques.