Claim Missing Document
Check
Articles

PENINGKATAN LITERASI CODING MELALUI PEMBELAJARAN BERBANTUAN AI SEBAGAI MEDIA INOVATIF Safrizal, Andrian Syahputra, Lili Tanti, Ramadani, Rabiana Saragih, Suci Sanjaya
JURNAL BIMA Vol 4 No 1 (2026): (Januari)
Publisher : PT.JNDI, JURNAL BIMA

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The development of digital technology and artificial intelligence (AI) demands innovation in the learning process, particularly in strengthening digital literacy and the computational thinking skills of teachers as learning designers. However, many educational institutions still face limitations in utilizing AI technology as a supporting tool for developing innovative and contextual learning media. This community service activity aims to provide training and mentoring to teachers in developing AI-assisted learning media through coding learning. The implementation method includes identifying partner needs, designing AI-based training materials, implementing training and mentoring directly, and evaluating the activity through observation and participant response questionnaires. The results of the activity show an increase in teachers' understanding and skills in utilizing AI for developing learning media, increased active participation during the training, and a positive response to the use of AI as an interactive and applicable learning medium. This activity has a positive impact on improving teachers' digital competence and becomes an alternative solution for developing innovative learning that can be implemented sustainably in educational environments.
Model Dynamic Facility Location in Post-Disaster Areas in Uncertainty lili Tanti; Syahril Efendi; Maya Silvi Lydia; Herman Mawengkang
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 22 No. 1 (2022)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v22i1.2095

Abstract

Indonesia has many disaster-prone areas, natural disasters that occur in Indonesia in 2021 are 5,402 disasters. For disaster management in post-disaster areas, logistical planning is needed in the distribution of logistical assistance, it is estimated that the logistics costs of disaster assistance reach approximately 80% of the total costs in disaster management so that logistical assistance is an expensive activity of disaster relief. However, so far the process of distributing logistical assistance to disaster posts has not been evenly distributed. One of the causes of the unequal distribution is the inappropriate selection of distribution post locations. The facility location model is dynamic and has the objective function of minimizing the distance between emergency posts and refugee posts in terms of distribution of disaster relief goods in one cluster group. For grouping unsupervised learning data using a machine learning clustering algorithm, k-means. Model validation has been carried out using max run and max optimization 1000 times with results reaching 90%. This proves that the emergency facility location model can be used to determine the location of the emergency center, where the determination of the location of the emergency center has the closest distance to the request point/post shelter for disaster victims
Algoritma Clustering untuk Menentukan Kategori Siswa Berdasarkan Prestasi pada SMA Pangeran Antasari Ayuni Syahputri; Lili Tanti
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 7 No. 2 (2025): Desember 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penentuan kategori prestasi siswa di SMA Pangeran Antasari masih dilakukan secara manual sehingga kurang efisien dan rentan menghasilkan penilaian yang subjektif. Penelitian ini menerapkan algoritma K-Means Clustering untuk mengelompokkan siswa ke dalam tiga kategori prestasi, yaitu tinggi, cukup, dan rendah. Dataset yang digunakan berjumlah 1.091 data siswa, mencakup nilai akademik, absensi, serta prestasi non-akademik. Proses evaluasi model dilakukan menggunakan metode Elbow untuk menentukan jumlah cluster optimal, dan pengujian menggunakan confusion matrix terhadap 30 data uji. Hasil penelitian menunjukkan bahwa algoritma K-Means mampu melakukan pengelompokan dengan baik, dengan tingkat akurasi sebesar 93,33%, serta rata-rata F1-score mencapai 94,2%. Temuan ini memperlihatkan bahwa K-Means efektif dalam memberikan pemetaan prestasi siswa secara objektif dan dapat membantu sekolah dalam proses evaluasi dan pengambilan keputusan akademik.
Model Klasifikasi Penyakit Kulit pada Kucing dengan Metode Machine Learning Alvian Julianto Hutajulu; Lili Tanti
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 7 No. 2 (2025): Desember 2025
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penelitian ini membahas pengembangan sistem klasifikasi penyakit kulit pada kucing menggunakan metode K-Nearest Neighbor (KNN). Latar belakang penelitian ini didasari oleh tingginya kasus penyakit kulit pada kucing serta keterbatasan akses terhadap layanan medis hewan saat klinik tidak tersedia. Sistem yang dirancang bertujuan untuk membantu pemilik kucing maupun dokter hewan dalam mendiagnosis penyakit berdasarkan gejala yang muncul. Data penelitian diperoleh melalui pengumpulan dataset gejala penyakit kulit kucing, kemudian dilakukan pra-pemrosesan data, transformasi, dan klasifikasi menggunakan algoritma KNN. Evaluasi dilakukan menggunakan confusion matrix dengan hasil akurasi sebesar 86%, presisi 80%, recall 86%, dan F1-score 83%. Hasil ini menunjukkan bahwa metode KNN cukup efektif dalam mengklasifikasikan jenis penyakit kulit kucing, seperti infeksi bakteri, alergi kulit, scabies, ringworm, dan feline acne. Sistem ini diharapkan dapat menjadi alternatif solusi dalam membantu pengambilan keputusan medis pada hewan peliharaan.
OPTIMIZING DECISION TREE PERFORMANCE WITH RECURSIVE FEATURE ELIMINATION FOR HIGH-DIMENSIONAL MUSHROOM CLASSIFICATION Lili Tanti; Safrizal; Yan Yang Thanri
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 2 (2025): JITK Issue November 2025
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i2.6816

Abstract

Classifying mushroom species presents a significant challenge within biological data analysis because of the wide variety of species and their distinct attributes. This research investigates the effectiveness of the Decision Tree classifier for mushroom categorization by comparing two splitting criteria, the Gini Index and Entropy. Additionally, the study employs the Recursive Feature Elimination (RFE) method for dimensionality reduction to enhance model efficiency and performance. The dataset was collected, cleaned, and analyzed exploratorily before feature selection was conducted using RFE. The Decision Tree model was trained and evaluated using accuracy, precision, recall, and F1-score metrics. The results showed that applying RFE improved computational efficiency without compromising model accuracy. The Gini criterion provided more stable results across all metrics, while Entropy demonstrated higher precision in certain cases. Model optimization through parameter tuning produced the best parameter combination at max_depth = 5, min_samples_leaf = 5, and min_samples_split = 10. This study concludes that integrating RFE with the Decision Tree can significantly enhance the performance of high-dimensional dataset classification. The findings are expected to serve as a reference for developing efficient and accurate biological data classification models
OPTIMIZATION OF MLP-NN FOR MANGO LEAF DISEASE PREDICTION USING IMAGE-BASED FEATURE EXTRACTION Budi Triandi; Lili Tanti
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 3 (2026): JITK Issue February 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i3.7031

Abstract

Mango (Mangifera indica Linn.) is a nutrient-rich fruit, yet leaf diseases caused by microorganisms can significantly reduce crop productivity. Early detection is essential to prevent further damage and support effective disease management. This study proposes an optimized mango leaf disease prediction model using a multi-layer perceptron neural network (MLP-NN). Image-based feature extraction is performed using the Inception v3 architecture to obtain high-level color and texture features that improve classification performance. Unlike previous studies that rely solely on manually engineered features or full CNN training, this research introduces a hybrid approach that integrates deep feature extraction with MLP-NN optimization, offering a lightweight yet highly accurate alternative. Several hyperparameter combinations, including activation functions (ReLU, tanh, sigmoid) and optimization algorithms (Adam and SGD), were evaluated using the Orange platform. The optimized MLP-NN model with ReLU and Adam achieved the highest accuracy of 93.5%, demonstrating better stability and training efficiency compared to other configurations. These findings highlight the novelty and advantages of the proposed method, showing improved accuracy with lower computational cost relative to many existing approaches. This study provides an efficient solution for mango leaf disease prediction and supports future development of automated plant disease detection systems
Optimasi Support Vector Machine Menggunakan Particle Swarm Optimization pada Analisis Sentimen Program Efisiensi Anggaran Pemerintah Nurhayati Nurhayati; Lili Tanti; Budi Triandi
Jurnal Minfo Polgan Vol. 15 No. 1 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i1.15929

Abstract

Respon publik terkait kebijakan pemerintah, termasuk program efisien anggaran dapat dipahami menggunakan pendekatan analisis sentimen. Penelitian ini bertujuan membandingkan kinerja algoritma Support Vector Machine (SVM) dengan kernel Radial Basis Functional (RBF), kernel Linear, dan kernel Polynomial, serta mengevaluasi pengaruh Particle Swarm Optimization (PSO) terhadap peningkatan performa klasifikasi. Dataset sebanyak 4274 data diperoleh melalui teknik crawling dari media sosial X (Twitter) dan kemudian diproses melalui langkah-langkah pembersihan teks seperti cleaning, case folding, normalization, tokenization, stopword removal, stemming, labeling menggunakan lexicon based serta menggunakan TF-IDF (Term Frequency-Inverse Document Frequency) untuk ekstraksi fitur. Proses penilaian kinerja model dilaksanakan dengan memanfaatkan indikator accuracy, precision, recall, dan F1-score, serta didukung oleh analisis confusion matrix. Berdasarkan hasil pengujian yang diperoleh, penelitian ini menunjukkan bahwa SVM kernel Linear berhasil meningkatkan akurasi menjadi 0.7579 atau 75.79%, sedangkan pada kernel RBF dan Polynomial tidak memberikan peningkatan signifikan. Selain itu, kelas netral menjadi kelas yang paling sulit diklasifikasikan. Penelitian ini menyimpulkan bahwa kombinasi SVM Linear dan PSO merupakan model terbaik untuk analisis sentimen kebijakan efisiensi anggaran pemerintah, serta menegaskan pentingnya pemilihan kernel dan strategi optimasi yang tepat dalam pengembangan sistem klasifikasi berbasis machine learning.
Penerapan Kombinasi Random Forest dan CNN untuk Mendeteksi Kecurangan Tagihan Listrik di Ulp Telda Albert Putra Nias Manao; Wanayumini Wanayumini; Lili Tanti
Jurnal Minfo Polgan Vol. 15 No. 1 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i1.16002

Abstract

Electricity fraud remains a major challenge for PLN (State Electricity Company), causing significant financial losses and reducing service reliability. This study aims to develop an electricity bill fraud detection system at PLN ULP Telda using a hybrid approach that combines Random Forest and Convolutional Neural Network (CNN) algorithms. Random Forest is effective in handling structured tabular data and identifying important features, while CNN excels in recognizing complex patterns in time-series electricity consumption data. By integrating both algorithms, the system is expected to achieve higher accuracy and reliability compared to a single model. The research methodology includes data collection from smart meters, pre-processing, feature extraction, model training, and evaluation using accuracy, precision, recall, and F1-score metrics. Initial results indicate that the hybrid model improves detection performance, reduces false negatives, and strengthens fraud identification. The final results show that the combined system outperforms the single model, with an accuracy of 91.3%, a precision of 89.7%, a recall of 88.4%, and an ROC-AUC of 90.5%. This research contributes to PLN's ability to detect fraud more accurately and efficiently. The contribution of this research is financial efficiency for PLN, fairer electricity tariffs for customers, and a reference for further research in the application of machine learning for energy fraud detection.
Peningkatan Keamanan Data Digital Dengan Pendekatan Kombinasi Algoritma Kriptografi OTP dan Cramer Shoup Gilang Dwi Fahri Harahap; Budi Triandi; Lili Tanti
Jurnal Minfo Polgan Vol. 15 No. 2 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i2.16080

Abstract

Keamanan data digital menjadi aspek krusial seiring meningkatnya ancaman siber terhadap kerahasiaan dan integritas informasi. Penelitian ini bertujuan untuk meningkatkan keamanan data digital melalui pendekatan kombinasi algoritma kriptografi One-Time Pad (OTP) dan Cramer-Shoup dalam sebuah skema hybrid. Algoritma OTP memiliki keunggulan dalam menjaga kerahasiaan data secara teoritis, namun memiliki kelemahan pada distribusi kunci dan potensi serangan seperti Known-Plaintext Attack (KPA). Sementara itu, algoritma Cramer-Shoup sebagai kriptografi asimetris menawarkan keamanan yang kuat terhadap serangan adaptif serta mendukung autentikasi data. Metode penelitian yang digunakan meliputi perancangan sistem hybrid, implementasi algoritma menggunakan bahasa pemrograman Python, serta pengujian keamanan dan kinerja sistem. Proses enkripsi dilakukan dengan mengombinasikan OTP untuk penyandian pesan dan Cramer-Shoup untuk pengamanan distribusi kunci. Evaluasi dilakukan melalui simulasi serangan KPA dan analisis perbandingan waktu proses enkripsi-dekripsi. Hasil penelitian menunjukkan bahwa kombinasi kedua algoritma mampu meningkatkan tingkat keamanan data secara signifikan, terutama dalam mengatasi kelemahan distribusi kunci pada OTP dan meningkatkan ketahanan terhadap serangan KPA. Selain itu, sistem hybrid yang diusulkan tetap mempertahankan efisiensi kinerja yang baik. Dengan demikian, pendekatan kombinasi OTP dan Cramer-Shoup dapat menjadi solusi efektif dalam meningkatkan keamanan data digital serta memberikan kontribusi dalam pengembangan sistem kriptografi yang lebih adaptif dan andal.
Optimalisasi Akurasi Naïve Bayes Menggunakan Seleksi Atribut Relief-F dan Gain Ratio Agung RM Alam; Wanayumini Wanayumini; Lili Tanti
Jurnal Minfo Polgan Vol. 15 No. 2 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i2.16112

Abstract

Naïve Bayes Classifier (NBC) merupakan salah satu algoritma klasifikasi probabilistik yang paling banyak digunakan dalam data mining karena kesederhanaan dan efisiensinya. Namun, performa NBC cenderung menurun ketika dataset mengandung atribut yang tidak relevan atau mengandung noise. Penelitian ini bertujuan menganalisis pengaruh seleksi atribut menggunakan metode Relief-F dan Gain Ratio terhadap peningkatan akurasi NBC. Dua dataset dari UCI Machine Learning Repository digunakan sebagai bahan pengujian: dataset House Vote (435 data, atribut simbolik) dan dataset Bank Marketing (45.211 data, atribut numerik dan kategorikal). Tiga skenario eksperimen diterapkan pada masing-masing dataset: (1) NBC tanpa seleksi atribut sebagai baseline, (2) NBC dengan seleksi atribut Relief-F, dan (3) NBC dengan seleksi atribut Gain Ratio. Evaluasi performa menggunakan 10-fold cross-validation dengan metrik akurasi, presisi, recall, F1-score, dan confusion matrix. Hasil penelitian menunjukkan bahwa pada dataset House Vote, Relief-F berhasil meningkatkan akurasi NBC dari 90,11% menjadi 93,79% (+3,68%), sedangkan Gain Ratio justru menurunkan akurasi menjadi 89,43%. Pada dataset Bank Marketing, Relief-F meningkatkan akurasi menjadi 89,36% dan memperbaiki recall kelas minoritas dari 29,34% menjadi 35,71%, sementara Gain Ratio hanya memberikan peningkatan marginal. Secara keseluruhan, Relief-F terbukti lebih efektif dibandingkan Gain Ratio dalam meningkatkan performa NBC, khususnya pada dataset dengan pola klasifikasi yang jelas dan distribusi kelas yang tidak seimbang.
Co-Authors adhar, Deni Adhar, Deni Adhar Agung RM Alam Ahmad, Ahmad Syah Lubis Ahsanul Huda Albert Putra Nias Manao Alim Murtani Alvian Julianto Hutajulu alya, Alya Rahmadani Andra Alfira Andra Alfitra Andrian Syahputra Anggi, Anggi Canita Simanjuntak Ayu Nadya Ayuni Syahputri Aziz Ritonga, Mirwan Bob Subahan Riza Bob Subhan Riza Bob Subhan Riza Bob Subhan Riza Bob Subhan Riza, Bob Subhan Budi Triandi Budi Triandi Budi Triandi, Budi Clarensia Mende Simalango Daifiria Daifiria Dedy Hartama Deni Adhar Deni Adhar Deni Adhar Adhar Deni Anggara Devi Pratiwi Putri Dewi Kartika Dhooni, Dhoni Briliant Dio Febrian Surbakti Efendi, Syahril Erica Rian Safitri Evri Ekadiansyah fachrie, Fachrie Ditya Faisal Tanjung Fauzan Arif Feberianus Zai Fhery Agustin Fretty S Siahaan Gilang Dwi Fahri Harahap Handoko, Muhammad Yan Handoko Putra F Hartama, Dedy Hartati Tammamah Lubis Herman Mawengkang Indah Widiastuti Iwan Fitrianto Rahmad Jasri, Jasri Ramadhan Juli Iriani Juli Iriani Juli Iriani Juli Iriani Iriani Juli Iriani, Juli Juni Ismail Khairul Fajri Khairul Ummi Khairun Dini Lahmudin Sipahutar Lubis, Hartati Tammamah Luthfi Zaidi Zaidi M Rizky M Zendi Lubis M. Aidil Fitra Wahyudi M. Haidil Umam Mangunsong, Puja Mas Ayoe Elhias Nasution Mas Ayoe Elhias Nst Mawardah Azzahra Maya Silvi Lydia Muhammad Daud Muhammad Faris Muhammad Suhada Nanda Setiawan Nety Juwita Lubis Nurainun Nurhayati Nurhayati Pairin, Yusfrizal Bin Patuan Putra Wijaya Sitorus Poningsih Poningsih Poningsih Poningsih, Poningsih Rabiana Saragih Ratih Puspasari Ratih Puspasari Ratih, Ratih Puspasari Rehan Nurli Pramudita Ginting Ridho, Irdian Riza, Bob Subahan Rofiqoh Dewi Romi Antoni Roslina Roslina Roslina Roslina Roslina Roslina, Roslina Safitri, Erica Rian Safrizal Safrizal Safrizal Safrizal Safrizal Safrizal Safrizal Safrizal Salwani D, Zuki Salwani D Susianto Susianto Syefira Arrafah Tasya Ardilah Thanri, Yan Yang Wanayumini Windy, Micky Alviansyah Wirhan Fahrozi Yanyang Thanri Yudhi Andrian Yulika Ababil Zulfa Khairunnisa _, Safrizal