cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Firdaus Annas
Contact Email
info@makwadfoundation.org
Phone
+6285278566869
Journal Mail Official
intellect.makwafoundation@gmail.com
Editorial Address
Jl. Dusun Pandam Jorong Aro Kandikir Nagari Gadut Kecamatan Tilatang Kamang Kabupaten Agam Sumatera Barat
Location
Kab. agam,
Sumatera barat
INDONESIA
Intellect : Indonesian Journal of Learning and Technological Innovation
Published by Yayasan Lembaga Studi Makwa
ISSN : -     EISSN : 29629233     DOI : -
Core Subject : Science, Education, Engineering,
Computer Science & IT Education Electrical & Electronics Engineering
The Intellect : Indonesian Journal of Learning and Technological Innovation aims to promote research and scholarship on the innovation of technology in secondary and higher education, as well as promote effective practice, and inform policy in education. The Intellect publishes papers related to theoretical foundations, design, analysis and implementation, as well as effectiveness and impact issues related to learning technology. The Intellect : Indonesian Journal of Learning and Technological Innovation published by Yayasan Lembaga Studi Makwa (Makwa Foundation)
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 110 Documents
 7   8   9   10   11 
Integrasi Canva dalam Pengembangan Media Video Pembelajaran Matematika Kelas IV SDIT Insan Kamil Muhammad Faishal; Liza Efriyanti
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1340

Abstract

The lack of creative ideas among teachers in designing mathematics learning media has led to low student interest and enthusiasm in learning mathematics at SDIT Insan Kamil Bukittinggi. This study aims to develop video-based learning media using the Canva application to assist teachers and students in the mathematics learning process. The research employed a Research and Development (R&D) model with the 4D approach, namely Define, Design, Develop, and Disseminate. The results show that the instructional videos uploaded to YouTube were validated by media experts, found practical by subject teachers, and proven effective and user-friendly for students. The findings confirm that this media increases students’ interest and engagement in mathematics learning. This research contributes by providing an innovative, valid, practical, and effective technology-based learning alternative to enhance the quality of mathematics education in elementary schools. Abstrak Kurangnya ide kreatif guru dalam merancang media pembelajaran matematika menyebabkan rendahnya minat dan antusiasme siswa dalam mengikuti pelajaran di SDIT Insan Kamil Bukittinggi. Penelitian ini bertujuan untuk mengembangkan media pembelajaran berbasis video menggunakan aplikasi Canva yang dapat membantu guru dan siswa dalam proses praktikum pembelajaran matematika. Model penelitian yang digunakan adalah Research and Development (R&D) dengan pendekatan 4D, yaitu Define, Design, Develop, dan Disseminate. Hasil penelitian menunjukkan bahwa video pembelajaran yang diunggah melalui platform YouTube terbukti valid menurut penilaian ahli media, praktis menurut guru mata pelajaran, serta efektif dan mudah digunakan oleh siswa. Temuan penelitian menegaskan bahwa media ini mampu meningkatkan ketertarikan dan keterlibatan siswa dalam pembelajaran matematika. Kontribusi penelitian ini adalah memberikan alternatif media pembelajaran berbasis teknologi yang inovatif, valid, praktis, dan efektif untuk mendukung peningkatan kualitas pendidikan matematika di sekolah dasar.
Development of an Integrated Raw Material Inventory Management Information System with a Food Menu for Profit and Loss Calculation Using the Rapid Application Development (RAD) Method (Case Study: Bento Kopi Pamulang) Dalia Oktaviyanti; Wasis Haryono
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 02 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i02.1361

Abstract

The use of information technology is a crucial factor in improving business efficiency, particularly in inventory management. Bento Kopi Pamulang still relies on manual recording of raw material inventory, which leads to various challenges such as data errors, information delays, and inaccurate COGS calculations. To address these challenges, this study developed a digital-based inventory management information system integrated with food menu data. The development was conducted using the Rapid Application Development (RAD) method, encompassing planning, design, prototype development, and implementation. Each stage was conducted iteratively with direct user involvement, from needs interviews and design validation to prototype trials and implementation evaluation. Active user involvement helped ensure the system met operational needs in the field. The resulting system was proven capable of recording raw material inflows and outflows in real time, automatically calculating COGS, and presenting faster and more accurate financial reports. System testing showed a 40% increase in recording efficiency and up to 95% increase in data accuracy compared to the manual system. Abstrak Pemanfaatan teknologi informasi menjadi faktor penting dalam meningkatkan efisiensi bisnis, terutama pada aspek pengelolaan persediaan. Bento Kopi Pamulang masih mengandalkan pencatatan manual untuk stok bahan baku, yang menyebabkan berbagai kendala seperti kesalahan data, keterlambatan informasi, dan perhitungan HPP yang tidak akurat. Untuk menjawab permasalahan tersebut, penelitian ini mengembangkan sistem informasi manajemen inventory berbasis digital yang terintegrasi dengan data menu makanan. Pengembangan dilakukan menggunakan metode Rapid Application Development (RAD), yang mencakup tahap perencanaan, desain, pengembangan prototipe, dan implementasi. Setiap tahap dilakukan secara iteratif dengan keterlibatan langsung dari pengguna, mulai dari wawancara kebutuhan, validasi desain, uji coba prototipe, hingga evaluasi implementasi. Keterlibatan aktif pengguna membantu memastikan sistem sesuai dengan kebutuhan operasional di lapangan. Sistem yang dihasilkan terbukti mampu mencatat keluar masuk bahan baku secara real-time, menghitung HPP secara otomatis, serta menyajikan laporan keuangan yang lebih cepat dan akurat. Pengujian sistem menunjukkan adanya peningkatan efisiensi waktu pencatatan sebesar 40%, dan akurasi data meningkat hingga 95% dibanding sistem manual.
Hegemoni Digitalisasi Pasar Tradisional di Era Revolusi 5.0 : Studi Kasus Pasar Atas Baru Cimahi Windri Wisti; Anggi Putra Kedua
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1362

Abstract

This article examines the phenomenon of the digitalization of traditional markets at Pasar Atas Cimahi in the era of Industry 5.0, utilizing Antonio Gramsci's theory of hegemony as an analytical framework. The research focuses on the implementation of the QRIS (Quick Response Code Indonesian Standard) cashless payment system. A qualitative method with a case study approach was employed, gathering data through policy documentation, media reports, and semi-structured interviews with market vendors. The findings indicate that despite administrative compliance with the digitalization policy, its implementation has not been optimal due to several obstacles, such as low digital literacy, a significant age gap, the necessity for cash flow, and the burden of transaction fees. From a Gramscian perspective, this reflects a form of hegemonyan ideological domination that is outwardly accepted but not fully internalized by the dominated group. Digitalization is not entirely perceived as progress by the vendors; instead, it is also seen as a new burden that does not align with their real-world needs.This study recommends a more inclusive and contextual policy approach to ensure that digital transformation genuinely addresses the interests and socio-economic conditions of lower-class communities in a fair and participatory manner. Abstrak Artikel ini membahas fenomena digitalisasi pasar tradisional di Pasar Atas Cimahi pada era Revolusi Industri 5.0 dengan menggunakan teori hegemoni Antonio Gramsci sebagai kerangka analisis. Penelitian ini berfokus dalam implementasi sistem pembayaran nontunai QRIS (Quick Response Code Indonesian Standard). Metode yang digunakan adalah kualitatif dengan pendekatan studi kasus, dengan teknik pengumpulan data melalui dokumentasi kebijakan, berita media, dan wawancara semi-terstruktur terhadap pedagang pasar. Hasil penelitian menunjukkan bahwa meskipun terdapat kepatuhan administratif terhadap kebijakan digitalisasi, implementasinya tidak berjalan optimal karena berbagai hambatan seperti rendahnya literasi digital, kesenjangan usia, kebutuhan perputaran uang tunai, dan beban biaya transaksi. Dalam perspektif Gramsci, hal ini mencerminkan bentuk hegemoni yaitu dominasi ideologis yang tampak diterima, tetapi belum dihayati secara penuh oleh kelompok yang didominasi. Digitalisasi tidak sepenuhnya dimaknai sebagai kemajuan oleh pedagang, melainkan juga sebagai beban baru yang tidak sesuai dengan kebutuhan riil mereka. Penelitian ini merekomendasikan pendekatan kebijakan yang lebih inklusif dan kontekstual, agar transformasi digital benar-benar menyentuh kepentingan dan kondisi sosial-ekonomi masyarakat lapisan bawah secara adil dan partisipatif.
A Review on Trends and Effectiveness of Rainfall Prediction Models for Smart Irrigation: Toward Future Development Olivia Wardhani; Rayfal Mayvandra Aurora Akbar Mayvandra Aurora Akbar; Yasabuana Athallahaufa Natawijaya
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1364

Abstract

Rainfall prediction is critical for enabling precision irrigation, particularly in tropical agricultural regions vulnerable to climate variability. This review systematically examines 15 peer-reviewed articles published between 2019 and 2024, using the PRISMA framework to evaluate the performance and applicability of rainfall prediction models for precision agriculture. The models are categorized into statistical (e.g., ARIMA), artificial intelligence (e.g., ANN, LSTM, ELM), and hybrid approaches (e.g., Neural Prophet–LSTM, ANFIS). Quantitative synthesis based on RMSE, MAE, MAPE, and R² reveals that hybrid models generally yield the highest predictive accuracy (e.g., RMSE = 0.0633; R² = 0.98), while AI models perform well on daily, nonlinear datasets but require extensive computational resources and expertise. In contrast, ARIMA remains the most practical and reliable option for monthly forecasting in data-scarce environments, offering a balance between accuracy and operational feasibility (e.g., RMSE = 69.506; MAPE = 31.41%). Contextual factors such as data availability, digital infrastructure, and user capacity significantly influence model suitability. The review also highlights real-world implementations and practical challenges—such as sensor limitations and technical skill gaps—associated with deploying advanced models. Ultimately, this review provides a comparative perspective to guide model selection based on statistical performance and implementation readiness. It further supports national food security goals by aligning predictive modeling with the operational needs of climate-resilient agriculture in supporting climate-resilient agriculture in tropical regions. Abstrak Prediksi curah hujan merupakan komponen penting dalam mendukung irigasi presisi, terutama di wilayah pertanian tropis yang rentan terhadap variabilitas iklim. Kajian ini secara sistematis menelaah 15 artikel ilmiah terbitan tahun 2019 hingga 2024 dengan menggunakan kerangka PRISMA, untuk mengevaluasi kinerja dan relevansi model prediksi curah hujan dalam konteks pertanian presisi. Model yang dianalisis mencakup pendekatan statistik (misalnya ARIMA), kecerdasan buatan (seperti ANN, LSTM, ELM), serta model hibrida (seperti Neural Prophet–LSTM dan ANFIS). Sintesis kuantitatif berdasarkan indikator RMSE, MAE, MAPE, dan R² menunjukkan bahwa model hibrida umumnya memberikan akurasi prediksi tertinggi (misalnya RMSE = 0,0633; R² = 0,98), sementara model AI efektif untuk data harian yang kompleks namun membutuhkan sumber daya komputasi dan keahlian teknis yang tinggi. Di sisi lain, ARIMA tetap menjadi pilihan paling praktis untuk peramalan bulanan di wilayah dengan keterbatasan data dan infrastruktur, karena mampu menyeimbangkan akurasi dan kemudahan operasional (misalnya RMSE = 69,506; MAPE = 31,41%). Faktor kontekstual seperti ketersediaan data, kesiapan infrastruktur digital, dan kapasitas pengguna sangat memengaruhi kesesuaian model. Kajian ini juga mengidentifikasi tantangan implementasi nyata, termasuk keterbatasan sensor dan rendahnya literasi teknologi. Secara keseluruhan, ulasan ini memberikan panduan komparatif dalam memilih model berdasarkan performa statistik dan kesiapan penerapan, serta mendukung upaya ketahanan pangan nasional melalui pemodelan prediksi yang kontekstual dan adaptif terhadap iklim.
Pengembangan Metode Sahl untuk Meningkatkan Pembelajaran Imla Aksara Arab Melayu di Madrasah Diniyah Takmiliyah Kecamatan Siabu Manahan Efendi; Zulhimma Zulhimma; Zainal Efendi Hasibuan
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1365

Abstract

The decline of the Arab-Malay academic tradition among students highlights the need for relevant and contextual learning innovations. This study aims to develop a learning model for Arab-Malay script to improve imla (dictation) learning outcomes in Madrasah Diniyah Takmiliyah (MDT) in Siabu District, Mandailing Natal Regency. The research employed the ADDIE development model, which includes the stages of needs analysis, design, development, implementation, and evaluation. The trial was conducted in three MDTs and showed that the developed Sahl method met the criteria of validity, practicality, and effectiveness. Students' learning outcomes improved in most madrasahs, and both teachers and students responded very positively regarding the practicality of the method. These findings indicate that the Sahl method is feasible for use in imla learning at MDTs as an effort to preserve the Arab-Malay cultural heritage. Abstrak Penurunan budaya akademik Arab Melayu di kalangan pelajar mendorong perlunya inovasi pembelajaran yang relevan dan kontekstual. Penelitian ini bertujuan untuk mengembangkan model pembelajaran aksara Arab Melayu guna meningkatkan hasil belajar imla di Madrasah Diniyah Takmiliyah (MDT) Kecamatan Siabu, Kabupaten Mandailing Natal. Metode yang digunakan adalah model pengembangan ADDIE, yang mencakup tahapan analisis kebutuhan, perancangan, pengembangan, implementasi, dan evaluasi. Uji coba dilaksanakan di tiga MDT, dan hasilnya menunjukkan bahwa metode Sahl yang dikembangkan memenuhi kriteria validitas, kepraktisan, dan keefektifan. Terdapat peningkatan hasil belajar siswa di sebagian besar madrasah, dengan respon guru dan siswa yang sangat positif terhadap kepraktisan metode. Temuan ini mengindikasikan bahwa metode Sahl layak diterapkan dalam pembelajaran imla di MDT sebagai upaya pelestarian budaya Arab Melayu.
Digitalisasi Pemesanan Produk Kerajinan Minangkabau Melalui Sistem Informasi Berbasis Web di UKM Putti Az-Zahra Dona Kurnia; Sadar Martua Haholongan Sir; Fauzan Azim
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1367

Abstract

Technological advancement has driven businesses to adopt digital systems to enhance efficiency and competitiveness, including within the small and medium enterprises (SMEs) sector. Putti Az-Zahra, an SME specializing in traditional Minangkabau handicrafts, still relies on manual processes for order management and reporting, which hampers operational effectiveness and market outreach. This study aims to design and develop a web-based ordering information system to improve transaction efficiency and expand customer access. The development method employed is the System Development Life Cycle (SDLC) using the waterfall model, consisting of requirement analysis, system design, implementation, testing, and maintenance stages. The result is a web-based information system that enables customers to place orders online and facilitates automatic transaction data management and sales reporting. The system was tested using the black box method, confirming that all functionalities meet user requirements. This research contributes to the digital transformation of ordering processes in traditional craft-based SMEs and serves as a model that can be adapted by similar enterprises to enhance competitiveness in the digital era. Abstrak Perkembangan teknologi mendorong perusahaan untuk meningkatkan efektivitas dan efisiensi melalui penerapan sistem digital, termasuk pada sektor Usaha Kecil dan Menengah (UKM). UKM Putti Az-Zahra, yang bergerak di bidang kerajinan tangan khas Minangkabau, menghadapi kendala dalam proses pemesanan dan pencatatan yang masih dilakukan secara manual, sehingga menghambat operasional dan jangkauan pasar. Penelitian ini bertujuan untuk merancang dan membangun sistem informasi pemesanan berbasis web guna meningkatkan efisiensi transaksi dan memperluas akses pelanggan. Metode pengembangan yang digunakan adalah System Development Life Cycle (SDLC) model waterfall, melalui tahapan analisis kebutuhan, perancangan sistem, implementasi, pengujian, dan pemeliharaan. Hasil dari penelitian ini adalah sistem informasi berbasis web yang memungkinkan konsumen melakukan pemesanan secara daring serta membantu pengelolaan data transaksi dan pelaporan penjualan secara otomatis. Pengujian menggunakan metode black box testing menunjukkan bahwa sistem telah berjalan sesuai fungsionalitas yang diharapkan. Penelitian ini memberikan kontribusi terhadap digitalisasi proses pemesanan pada sektor UKM kerajinan tradisional dan dapat menjadi model yang diadaptasi oleh UKM lain dalam meningkatkan daya saing di era digital.
Monitoring Real-time Server Dan Router Berbasis Zabbix Dengan Notifikasi Alert Telegram: Studi Pra-Pasca Di SDIT Raudhatul As-Salimy Gobah Putri Dwi Rahayu; Andi Ahmad Dahlan; Rikki Vitria
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1372

Abstract

Reliable network monitoring technology is essential for educational institutions to ensure uninterrupted digital services. This study aims to implement a Zabbix-based server and router monitoring system integrated with Telegram notifications at SDIT Raudhatul As-Salimy Gobah. A mixed-methods approach was applied, combining quantitative pre–post evaluation with qualitative interviews and observations of the school’s IT staff. The results show that the system detected failures with an average delay of 2 minutes 30 seconds and delivered recovery notifications within 20 seconds. Integration with Telegram significantly reduced the mean time to acknowledge (MTTA) from 1–2 hours to 10–20 minutes, thereby improving IT team responsiveness in handling incidents. This implementation not only minimized network downtime but also enhanced operational efficiency, infrastructure stability, and the quality of digital-based learning services. The main contribution of this study is the development of a low-cost, open-source, and replicable network monitoring model suitable for K-12 schools in rural areas. Abstrak Pemanfaatan teknologi monitoring jaringan yang handal menjadi kebutuhan penting bagi institusi pendidikan untuk menjaga kelancaran layanan digital. Penelitian ini bertujuan mengimplementasikan sistem monitoring server dan router berbasis Zabbix yang terintegrasi dengan notifikasi Telegram di SDIT Raudhatul As-Salimy Gobah. Metode penelitian menggunakan pendekatan mixed-methods, yaitu evaluasi kuantitatif pre–post implementasi serta wawancara dan observasi kualitatif terhadap tim IT sekolah. Hasil pengujian menunjukkan bahwa sistem mampu mendeteksi gangguan dengan rata-rata waktu 2 menit 30 detik dan mengirimkan notifikasi dalam 20 detik setelah perangkat kembali normal. Integrasi dengan Telegram secara signifikan menurunkan mean time to acknowledge (MTTA) dari 1–2 jam menjadi 10–20 menit, serta memperbaiki responsivitas tim IT dalam menangani gangguan. Implementasi ini tidak hanya mengurangi downtime jaringan, tetapi juga meningkatkan efisiensi kerja, stabilitas infrastruktur, serta kualitas layanan pembelajaran berbasis digital. Kontribusi penelitian ini adalah penyusunan model monitoring jaringan open-source, biaya rendah, dan replikatif yang dapat diterapkan oleh sekolah K-12 di wilayah rural.
Implementasi Algoritma Decision Tree dengan Pruning pada Sistem Manajemen Data Jemaat: Studi Kasus GKS Umamapu Pius Tae Here Ati; Rambu Yetti Kalaway; Erwiantah Gustial Radjah
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1374

Abstract

The church plays an important role in managing congregation data, including membership records, sacrament administration, and monitoring congregation development. At GKS Jemaat Umamapu, data management is still carried out manually, which poses risks such as data loss, recording errors, and delays in accessing information. This study aims to develop a web-based congregation management system to improve administrative efficiency and accuracy. The system was designed using the Rapid Application Development (RAD) method with direct user involvement at each stage. A Decision Tree algorithm with pruning techniques was applied to identify members eligible for catechesis while reducing model complexity and the risk of overfitting. The results of Black Box testing showed that all system functions operated according to user requirements without significant errors. The implementation of the Decision Tree with pruning achieved a classification accuracy of 92%, which is higher than the model without pruning. In addition, the system successfully reduced the average processing time of administrative tasks by approximately 40% compared to manual methods. These findings indicate that the developed system is not only valid and reliable but also efficient in supporting data-driven decision-making. This research contributes by providing an innovative technology-based congregation management system that strengthens church administration and enhances pastoral services sustainably in the digital era. Abstrak Gereja memiliki peran penting dalam pengelolaan data jemaat, seperti pencatatan keanggotaan, administrasi sakramen, dan pemantauan perkembangan jemaat. Di GKS Jemaat Umamapu, pengelolaan data masih dilakukan secara manual, sehingga berisiko menyebabkan kehilangan data, kesalahan pencatatan, dan keterlambatan akses informasi. Penelitian ini bertujuan mengembangkan sistem pengelolaan data jemaat berbasis web untuk meningkatkan efisiensi dan akurasi administrasi. Sistem dirancang menggunakan metode Rapid Application Development (RAD) dengan melibatkan masukan langsung dari pengguna di setiap tahap. Algoritma Decision Tree dengan teknik pruning diterapkan untuk mengidentifikasi jemaat yang layak mengikuti katekisasi, sekaligus mengurangi kompleksitas dan risiko overfitting.  Hasil pengujian menggunakan metode Black Box menunjukkan bahwa seluruh fungsi sistem berjalan sesuai kebutuhan pengguna tanpa error yang signifikan. Implementasi algoritma Decision Tree dengan pruning mampu meningkatkan akurasi klasifikasi jemaat hingga 92%, lebih baik dibandingkan tanpa pruning. Selain itu, sistem ini terbukti mampu mengurangi waktu pengolahan data administrasi rata-rata hingga 40% lebih cepat dibanding metode manual. Temuan ini menegaskan bahwa sistem yang dikembangkan tidak hanya valid dan andal, tetapi juga efisien dalam mendukung pengambilan keputusan berbasis data. Penelitian ini berkontribusi dalam menyediakan inovasi sistem manajemen jemaat berbasis teknologi yang dapat memperkuat struktur administrasi gereja dan meningkatkan layanan pastoral secara berkelanjutan di era digital.
Analisis Sentimen Komentar YouTube pada Program Clash of Champions Ruangguru Menggunakan Deep Learning Berbasis LSTM Iska Rahmawati; Nur Aini
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1376

Abstract

The growth of digital platforms such as YouTube has opened massive interactive spaces among users, particularly through comment sections on educational content. This study aims to analyze public sentiment toward Ruangguru's "Clash of Champions" program by utilizing user comments on YouTube and applying the Long Short-Term Memory (LSTM) algorithm. A total of 43,664 comments were collected over 46 days using the YouTube API, then processed through several stages: data preprocessing (text cleaning, tokenizing, stemming), manual sentiment labeling, feature transformation using TF-IDF, and LSTM model training that was fine-tuned with informal Indonesian language data. The classification results show that 72.6% of the comments carry positive sentiment, while 27.4% are negative. Model evaluation using accuracy, precision, recall, and F1-score yielded values of 90%, indicating a stable and reliable classification performance. These findings demonstrate the effectiveness of LSTM in capturing sentiment patterns within informal and local language texts and contribute by presenting a deep learning-based approach to analyze public opinion toward educational digital content in Indonesia. Abstrak Pertumbuhan platform digital seperti YouTube telah membuka ruang interaksi yang masif antar pengguna, khususnya melalui kolom komentar pada konten edukatif. Penelitian ini bertujuan untuk menganalisis sentimen publik terhadap program "Clash of Champions" dari Ruangguru dengan memanfaatkan komentar pengguna di YouTube dan menggunakan algoritma Long Short-Term Memory (LSTM). Sebanyak 43.664 komentar dikumpulkan selama 46 hari menggunakan YouTube API, kemudian diproses melalui tahap data preprocessing (text cleaning, tokenizing, stemming), pelabelan sentimen secara manual, transformasi fitur menggunakan TF-IDF, dan pelatihan model LSTM yang telah difine-tuning dengan data relevan berbahasa Indonesia informal. Hasil klasifikasi menunjukkan bahwa 72,6% komentar bersentimen positif dan 27,4% bersentimen negatif. Evaluasi model menggunakan metrik accuracy, precision, recall, dan F1-score dengan nilai mencapai 90%, menunjukkan bahwa model memiliki kemampuan klasifikasi yang stabil dan andal. Temuan ini menunjukkan efektivitas LSTM dalam menangkap pola sentimen pada data teks yang bersifat informal dan lokal, serta memberikan kontribusi dengan menyajikan pendekatan berbasis deep learning untuk analisis opini publik terhadap konten edukatif digital di Indonesia.
Analisis Perbandingan Prediksi Harga Rumah Dengan Random Forest, Gradient Boosting, dan XGBoost Bety Wulan Sari; Donni Prabowo
Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation
Publisher : Yayasan Lembaga Studi Makwa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57255/intellect.v4i1.1385

Abstract

House price prediction poses a significant challenge in the property sector, especially in the Yogyakarta region, which exhibits a wide range of price variations. This study aims to compare the performance of three regression algorithms such as Random Forest, Gradient Boosting, and XGBoost, in building predictive models based on features such as land area, building area, number of bedrooms, bathrooms, and garage availability. The dataset analyzed consists of 1,642 entries, with house prices ranging from IDR 7 million to IDR 4.37 billion, an average price of IDR 1.14 billion, and a mode of IDR 775 million. Model evaluation was conducted using Mean Squared Error (MSE) and the coefficient of determination (R²), where XGBoost achieved the best performance with an MSE of 1.56 × 10¹⁴ IDR², an R² of 0.7746, and a Root Mean Squared Error (RMSE) of approximately IDR 12.5 million. These results indicate that XGBoost outperforms the other two models in handling complex tabular data and provides more accurate predictions. The predictive model has practical potential to be utilized by property developers, real estate agents, and local governments as a decision-support tool for price estimation, market evaluation, and data-driven urban planning. These findings highlight that selecting the appropriate algorithm can significantly enhance the quality of house price prediction. Abstrak Prediksi harga rumah menjadi tantangan penting dalam bidang properti, khususnya di wilayah Yogyakarta yang memiliki variasi harga cukup ekstrem. Penelitian ini bertujuan untuk membandingkan performa tiga algoritma regresi yaitu Random Forest, Gradient Boosting, dan XGBoost digunakan untuk membangun model prediksi harga rumah berdasarkan fitur seperti luas tanah, luas bangunan, jumlah kamar tidur, kamar mandi, dan garasi. Data yang dianalisis mencakup 1.642 entri dengan harga rumah berkisar antara Rp 7 juta hingga Rp 4,37 miliar, harga rata-rata sebesar Rp 1,14 miliar, dan modus Rp 775 juta. Evaluasi model dilakukan menggunakan metrik Mean Squared Error (MSE) dan koefisien determinasi (R²), di mana XGBoost menghasilkan performa terbaik dengan MSE sebesar 1,56 × 10¹⁴ rupiah², R² sebesar 0,7746, dan Root Mean Squared Error (RMSE) sekitar 12,5 juta rupiah. Hasil ini menunjukkan bahwa XGBoost lebih unggul dalam menangani data tabular kompleks dan memiliki akurasi prediksi yang lebih baik dibanding dua model lainnya. Model prediktif ini berpotensi digunakan oleh pengembang properti, agen real estate, maupun pemerintah daerah sebagai alat bantu dalam penetapan harga, evaluasi pasar, dan perencanaan tata ruang yang berbasis data. Temuan ini memberikan gambaran bahwa pemilihan algoritma yang tepat dapat meningkatkan kualitas prediksi harga properti.

Page 9 of 11 | Total Record : 110

 7   8   9   10   11 

Filter by Year

2022 2026


Reset
Filter By Issues
All Issue Vol. 5 No. 01 (2026): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 02 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 4 No. 1 (2025): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 3 No. 2 (2024): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 3 No. 1 (2024): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 2 No. 2 (2023): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 2 No. 1 (2023): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 1 No. 2 (2022): Intellect : Indonesian Journal of Learning and Technological Innovation Vol. 1 No. 1 (2022): Intellect : Indonesian Journal of Learning and Technological Innovation