Abstract:The number of rail passengers on the Kisaran–Medan route varies over time due to a number of variables, including local economic situations, national holidays, peak travel seasons, and current transportation regulations.…
s. The administration of Kisaran Station finds it challenging to manage train fleets, organize travel timetables, and enhance service quality due to this unpredictability. Using historical data from 2019 to 2025, this study intends to apply an Artificial Neural Network (ANN) with the Backpropagation technique to forecast the number of train passengers at Kisaran Station. With a learning rate of 0.25 and a sigmoid binary activation function, the system was constructed using a 7-5-1 network architecture with seven input neurons (yearly data from 2019–2025), five hidden layer neurons, and one output neuron. The data was split into 70% training data (January–July) and 30% testing data (August–December) after being adjusted to the interval 0.1–0.9 using the Min-Max technique. At epoch 822, the training process reached a convergence. For most months, including March (predicted: 647,271; actual: 642,870 passengers) and April (predicted: 999,728; actual: 996,320 passengers), the model was able to produce predictions that were close to actual values, according to testing results. However, there were differences in some months with seasonal spikes. It is anticipated that PT Kereta Api Indonesia and the regional government would use the created prediction system as a decision-support tool to plan transportation services and enhance the general quality of railway services.
Abstract:The Free Nutritional Meal Program (MBG) is a government effort to improve the nutritional quality of students by providing nutritious food. This study aims to analyze the prediction of nutrient consumption levels in the…
MBG Program using the K-Nearest Neighbor (KNN) method. The data used were 100 datasets consisting of variables such as age, menu type, menu calories, initial nutritional status, and nutrient consumption levels. The data were divided into 80% training data and 20% testing data with a K value of 5. The test results showed that the KNN method achieved 95% accuracy, with 19 data points correctly predicted out of 20 test data points. The results indicate that the KNN method is capable of predicting nutrient consumption levels effectively and can be an alternative to support the evaluation of the Free Nutritional Meal Program (MBG).
Abstract:The development of artificial intelligence (AI)-based technology has encouraged organizations to adopt digital systems for human resource management, one of which is through a cortex-based system. This study aims to analyze…
yze the influence of cortex on the effectiveness of employee performance in managing human resources. The research method used was a qualitative case study approach. Data were collected through in-depth interviews, observation, and documentation involving 10 informants. The results show that cortex has a positive influence on the effectiveness of employee performance, particularly in improving efficiency, data accuracy, and work productivity. However, this influence is not optimal due to obstacles such as limited digital competency, lack of training, and resistance to change. Therefore, organizations need to strengthen human resource readiness and adaptive strategies to optimize the implementation of the cortex systemCortex
Abstract:Abstract: The buffer zone of the Nantu-Boliyohuto Wildlife Sanctuary in Boalemo, Gorontalo, is a tropical rainforest harboring rich avifauna biodiversity for conservation-based ecotourism. This study identified bird species…
ies with ecotourism potential and determined their diversity level. Conducted in November 2025 in Saritani Village, the research employed a quantitative descriptive method with an exploratory approach across three zones: forest edge, riparian, and agroforestry. Results revealed 20 bird species from 17 families, totaling 198 individuals. The overall diversity index was moderate ( = 2.0746) with Dicaeum celebicum (Grey-sided Flowerpecker) being the most frequently encountered. Out of these, 10 species possess significant ecotourism potential based on striking plumage, distinctive birdsong, endemism, conservation status, and unique behavior. Key species include the Knobbed Hornbill (Rhyticeros cassidix), a Vulnerable endemic species, the Small Blue Kingfisher (Alcedo coerulescens), the Collared Kingfisher (Todirhamphus chloris), the Black Kite (Milvus migrans), and the Javan/Pied Myna (Gracupica contra). These findings indicate that the Nantu-Boliyohuto buffer zone holds substantial potential for development as a sustainable birdwatching (avitourism) destination. This wealth of biodiversity can serve as a primary attraction while simultaneously supporting regional conservation efforts and local community awareness.
Keywords: Avifauna, Birdwatching, Ecotourism, Buffer Zone, Nantu-Boliyohuto Wildlife Sanctuary.
Abstrak: Kawasan penyangga Suaka Margastwa Nantu-Boliyohuto merupakan kawasan transisi hutan hujan tropis di Kabupaten Boalemo, Gorontalo, yang menyimpan potensi keanekaragaman hayati, khusunya avifauna, sebagai daya tarik ekowisata berbasis konservasi. Penelitian ini bertujuan untuk mengidentifikasi jenis-jenis burung yang berpotensi menjadi daya tarik ekowisata serta mengetahui tingkat keanekaragaman jenis burung di kawasan tersebut. Penelitian ini dilaksanakan pada bulan November 2025 di Desa Saritani, Kecamatan Wonosari, Kabupaten Boalemo, menggunakan metode Deskriptif Kuantitatif dengan pendekatan eksplorasi pada tiga titik pengamatan, yaitu jalur tepi hutan, area riparian, dan kawasan agroforestri. Hasil penelitian ditemukan 20 spesies burung dari 17 famili dengan total 198 individu. Indeks keanekaragaman secara keseluruhan termasuk dalam kategori sedang ( = 2.0746), dengan spesies yang paling banyak dijumpai adalah Dicaeum celebicum (Cabai panggul merah/kelabu). Dari 20 spesies yang teridentifikasi, terdapat 10 spesies yang berpotensi menjadi daya tarik ekowisata berdasarkan kriteria warna bulu mencolok, suara kicauan khas, status endemik, status konservasi, dan perilaku unik, diantaranya Julang Sulawesi (Rhyticeros cassidix) sebagai spesies endemik Vulnerable, Raja Udang Biru (Alcedo coerulescens), Cekakak Sungai (Todirhamphus chloris), Elang Paria (Milvus migrans), dan Jalak Suren (Gracupica contra). Hal ini menunjukkan bahwa kawasan penyangga Suaka Margasatwa Nantu-Boliyohuto memiliki potensi yang signifikan untuk dikembangkan sebagai destinasi ekowisata berbasis pengamatan burung (birdwatching/avitourism) yang berkelanjutan.
Kata kunci: Avifauna, Birdwatching, Ekowisata, Kawasan Penyangga, Suaka Margasatwa Nantu-Boliyohuto.
Abstract:Abstract : The development of information technology and social media has made it easier for people to express their opinions on various issues, including the government’s responsiveness in handling flood disasters. The…
e large number of comments generated on social media makes manual opinion analysis ineffective because it requires a long time and involves a large volume of data. Therefore, an automated method is needed to process and classify public sentiment. This study aims to implement text mining to analyze public sentiment toward the government’s responsiveness in flood disaster management using the K-Nearest Neighbour (KNN) algorithm. The research data were collected from social media platforms such as Instagram, TikTok, and X, consisting of public comments related to flood disasters in Aceh and North Sumatra. The research stages include data collection, text preprocessing, feature extraction, data splitting into training and testing sets, sentiment classification using the KNN method, and evaluation of classification results. The system was developed as a web-based application using PHP programming language and MySQL database. The results of this study are expected to provide an overview of public opinion tendencies in the form of positive, negative, and neutral sentiments and serve as evaluation material for the government to improve its response in flood disaster management.
Keywords: Text Mining, Sentiment Analysis, K-Nearest Neighbour (KNN), Social Media, Flood Disaster, Government Responsiveness.
Abstrak : Perkembangan teknologi informasi dan media sosial menyebabkan masyarakat semakin mudah menyampaikan pendapat terhadap berbagai isu, termasuk ketanggapan pemerintah dalam penanggulangan bencana banjir. Banyaknya komentar yang muncul di media sosial menjadikan analisis opini secara manual tidak efektif karena membutuhkan waktu yang lama dan jumlah data yang besar. Oleh karena itu, diperlukan metode otomatis untuk mengolah dan mengklasifikasikan sentimen masyarakat. Penelitian ini bertujuan untuk mengimplementasikan text mining dalam menganalisis sentimen masyarakat terhadap ketanggapan pemerintah dalam penanggulangan bencana banjir menggunakan algoritma K-Nearest Neighbour (KNN). Data penelitian diperoleh dari platform media sosial Instagram, TikTok, dan X yang berisi komentar masyarakat terkait bencana banjir di wilayah Aceh dan Sumatera Utara. Tahapan penelitian meliputi pengumpulan data, preprocessing teks, ekstraksi fitur, pembagian data training dan testing, proses klasifikasi menggunakan metode KNN, serta evaluasi hasil klasifikasi. Sistem dibangun berbasis web menggunakan bahasa pemrograman PHP dan database MySQL. Hasil penelitian diharapkan dapat memberikan gambaran kecenderungan opini masyarakat dalam bentuk sentimen positif, negatif, dan netral serta membantu sebagai bahan evaluasi bagi pemerintah dalam meningkatkan kualitas respon terhadap bencana banjir.
Kata kunci: : Text Mining, Analisis Sentimen, K-Nearest Neighbour (KNN), Media Sosial, Bencana Banjir, Ketanggapan Pemerintah.
Abstract:Abstract: The availability of clean water is a basic need that significantly impacts the health and quality of life of the community, especially in rural areas such as Sei Kepayang District, which still relies on well water…
ter and surface air sources with quality that does not always meet health standards, such as high turbidity levels and unsuitable air pH. The need for clean water for the Indonesian people is very important. Whether it is used for drinking water, cooking, bathing and washing. However, this need for clean water is very difficult to meet in areas far from water sources, especially private wells. Most air sources obtained from wells are often cloudy and the water pH does not match the nominal limit. One example is in Sei Kepayang District. Many residents still use rain-fed wells whose water quality is not always good. Given this problem, a solution that can be applied is to monitor air quality to determine the good or bad air quality so that it can be used by the community. The implementation of IoT (internet of things) air quality monitoring and administration systems in small-scale clean water management that displays conditions visually on the monitoring feature can be controlled efficiently with wireless media via the website. Furthermore, the addition of a sensor to detect and filter air pH levels can help communities struggling to access clean air and determine whether it is fit for consumption. The monitoring website also displays monthly PDAM usage costs, obtained from flow meter sensor data. This development falls under small-scale Smart City management, as it helps communities modernize their daily appliances. The implementation of the IoT (Internet of Things) air quality monitoring and administration system involved testing each sensor. Tests were conducted on each device. The test results showed that the pH sensor obtained a value of 7.00 for mineral water, 5.9 for lemon juice, and 10.4 for soapy water, and a litmus test for acid-base determination. Tests on the turbidity sensor for clean water yielded an average error of 0.12%, 0.02% for tea, and 0.08% for coffee.
Keywords: Internet of Things, air quality monitoring, Administration, pH meter, turbidity, flow meter, Arduino uno, Smart City, ESP8266.
Abstrak: Ketersediaan air bersih merupakan kebutuhan mendasar yang sangat berpengaruh terhadap kesehatan dan kualitas hidup masyarakat, khususnya di wilayah pedesaan seperti Kecamatan Sei Kepayang yang masih bergantung pada sumber air sumur dan air permukaan dengan kualitas yang tidak selalu memenuhi standar kesehatan, seperti tingkat kekeruhan yang tinggi dan pH air yang tidak sesuai. Kebutuhan air bersih untuk masyarakat Indonesia sangatlah penting. Baik itu digunakan untuk air minum , memasak, mandi dan mencuci. Namun kebutuhan air bersih ini sangat sulit di penuhi di daerah yang jauh dari sumber mata air, terlebih lagi sumur-sumur milik pribadi. Kebanyakan sumber air yang diperoleh dari sumur sering sekali keruh dan pH airnya tidak sesuai dari batas nomal. Salah satunya di Kecamatan Sei Kepayang. Masih banyak penduduknya menggunakan sumur tadah hujan yang kualitas airnya tidak selalu baik. Dengan adanya masalah tersebut, solusi yang dapat diterapkan adalah monitoring kualitas air untuk menentukan baik buruknya kualitas air agar bisa digunakan oleh masyarakat. Dikembangkan implementasi IoT (internet of things) monitoring kualitas air dan sistem administrasi pada pengelola air bersih skala kecil yang menampilkan keadaan secara visual pada fitur monitoring dapat dikendalikan secara efisien dengan media wireless melalui website. Disamping itu dengan ditambahkannya sebuah sensor untuk mendeteksi kadar pH air dan penyaringannya dapat membantu penduduk yang sulit mendapatkan air bersih maupun menentukan bahwa air tersebut layak tidaknya untuk di konsumsi. Serta menampilkan biaya penggunaan PDAM dalam kurun waktu bulanan pada website monitoring yang didapatkan dari data sensor flow meter. Pengembangan ini termasuk dalam pengelolaan Smart City skala kecil, karena pengembangan alat ini membantu penduduk untuk lebih moderenisasi alat-alat dalam kehidupan sehari- hari . Dari Implementasi IoT (internet of things) monitoring kualitas air dan sistem administrasi dilakukan pengujian pada Setiap alat yang dilakukan pada setiap sensor yang di gunakan pada tiap alat. Diperoleh Hasil pengujian menunjukkan bahwa sensor pH didapatkan nilai 7.00 untuk air mineral, air lemon 5.9, air sabun 10.4, dan dengan indikasi lakmus dalam penentuan asam basa pada air. Pengujian diperoleh pada sensor turbidity terhadap air bersih didapatkan nilai rata-rata error 0.12%, teh 0.02%, dan air kopi 0.08%.
Kata kunci : Internet of Things , monitoring kualitas air, Administrasi,pH meter, turbidity,flow meter , Arduino
uno,Smart City,ESP8266.
Abstract:Extreme weather in coastal areas such as Kota Padang is influenced by complex topography and local atmospheric dynamics, making daily rainfall prediction a critical challenge for disaster mitigation and sectoral planning.…
. This study aims to develop a daily rainfall prediction model using a Deep Neural Network (DNN) and evaluate its performance. The dataset consists of daily temperature, humidity, and air pressure data collected from the Automatic Weather Station (AWS) of BMKG Maritime Teluk Bayur from January 2020 to December 2024, comprising 1,823 samples after preprocessing. The research methodology follows the CRISP-DM framework, which includes six main phases. The proposed DNN architecture contains three hidden layers (128, 64, and 32 neurons) with ReLU activation and dropout regularization, enhanced by lag and rolling average features to capture temporal patterns. The model achieved an MAE of 1.79, RMSE of 17.31, and R² of 0.8972, indicating strong predictive performance. The model was deployed through a Streamlit-based interactive dashboard
Abstract:This study aims to analyze and classify student graduation based on academic data using the Artificial Neural Network (ANN) method with the Perceptron algorithm. The background of this research is the low rate of on-time…
student graduation caused by differences in students’ academic achievements during their studies. The academic data used in this research include Grade Point Average (GPA), total passed credits, and total failed credits. The study was conducted using data from Informatics Engineering students of Universitas Asahan class of 2021.The research method used is a quantitative method with stages including data preprocessing, classification target determination, weight initialization, perceptron training process, activation function calculation, error evaluation, and weight updating until convergence is achieved. The dataset was divided into training data and testing data to evaluate the model’s ability to classify student graduation into two categories, namely on-time graduation and delayed graduation. The system was developed using the PHP programming language and MySQL database and designed using Unified Modeling Language (UML).The results of this study indicate that the Artificial Neural Network method with the Perceptron algorithm is capable of classifying student graduation based on academic data effectively. The perceptron model is able to recognize the relationship patterns between GPA, passed credits, and failed credits variables toward student graduation status. The developed system can also assist the study program in academic evaluation and decision-making processes related to improving the quality of student graduation.
Abstract:Abstract: English language training for elementary school students is one way to improve basic communication skills in a foreign language from an early age. This study was conducted to explore the role-play method in teaching…
ching English to elementary school students, which aims to create an interactive, enjoyable, and student centred learning environment. Through this method, students are given the opportunity to act out various simple situations frequently encountered in everyday life, such as introducing oneself, greeting, asking and giving information, and shopping conversations. The training was carried out in stages through the provision of vocabulary, example dialogues, group exercises, and role-playing in front of the class. The training results showed that the role-play method was able to increase students' confidence in speaking English, enrich vocabulary mastery, and improve pronunciation and understanding of the context of language use. In addition, students showed high enthusiasm and participation throughout the activity. Thus, the role-play method can be an effective learning alternative to support the development of English language skills in elementary school students.
Keywords: English language training, elementary school, role play
Abstrak: Pelatihan bahasa Inggris bagi siswa sekolah dasar merupakan salah satu upaya untuk meningkatkan kemampuan komunikasi dasar dalam bahasa asing sejak usia dini. Penelitian ini dilaksanakan untuk mengksplorasi metode role play dalam mengajar bahasa Inggris bagi siswa sekolah dasar yang bertujuan menciptakan suasana belajar yang interaktif, menyenangkan, dan berpusat pada siswa. Melalui metode ini, siswa diberikan kesempatan untuk memerankan berbagai situasi sederhana yang sering ditemui dalam kehidupan sehari-hari, seperti introducing oneself, greeting, asking and giving information, dan shopping conversation. Pelatihan dilaksanakan secara bertahap melalui pemberian kosakata, contoh dialog, latihan kelompok, dan penampilan peran di depan kelas. Hasil pelatihan menunjukkan bahwa metode role play mampu meningkatkan kepercayaan diri siswa dalam berbicara bahasa Inggris, memperkaya penguasaan kosakata, serta meningkatkan kemampuan pelafalan dan pemahaman konteks penggunaan bahasa. Selain itu, siswa menunjukkan antusiasme dan partisipasi yang tinggi selama kegiatan berlangsung. Dengan demikian, metode role play dapat menjadi alternatif pembelajaran yang efektif untuk mendukung pengembangan keterampilan berbahasa Inggris pada siswa sekolah dasar.
Kata kunci: pelatihan bahasa Inggris, sekolah dasar, role play.
Abstract:Abstract: Child developmental disorders are conditions that may affect children's cognitive, behavioral, social, and communication abilities. Early identification of developmental disorders is essential to ensure timely…
intervention and appropriate treatment. However, the classification of developmental disorders remains challenging due to the similarities of symptoms among different disorders. Therefore, this study aims to implement the Support Vector Machine (SVM) algorithm for the classification of child developmental disorders, specifically ADHD, Autism Spectrum Disorder and Down Syndrome. This study employed a quantitative approach using a dataset consisting of 1,786 records with nine predictor attributes related to developmental symptoms. Data preprocessing was performed through missing value handling using the Most Frequent Imputation method and label encoding. The dataset was divided into training and testing sets using an 80:20 ratio. The SVM model was developed using the Radial Basis Function (RBF) kernel and evaluated using a confusion matrix, accuracy, precision, recall, and F1-score metrics. The experimental results demonstrated that the proposed SVM model achieved an accuracy of 98.88%. The classification report showed high performance across all classes, with precision, recall, and F1-score values reaching 0.99 on average. Confusion matrix analysis indicated only four misclassified instances out of 358 testing samples, while the Down Syndrome class achieved perfect classification performance. The findings indicate that the SVM algorithm is highly effective for classifying child developmental disorders and has the potential to support early identification and decision-making processes in child healthcare. The proposed model can serve as a reliable decision-support tool for healthcare professionals and parents in detecting developmental disorders based on observed symptoms.
Keywords: SVM, Machine Learning, Developmental Disorders, ADHD, Autism, Down Syndrome, Classification.
Abstrak: Gangguan perkembangan anak merupakan kondisi yang dapat memengaruhi kemampuan kognitif, perilaku, sosial, dan komunikasi anak. Identifikasi dini terhadap gangguan perkembangan sangat penting untuk memastikan intervensi yang tepat waktu dan penanganan yang sesuai. Namun, klasifikasi gangguan perkembangan masih menjadi tantangan karena adanya kemiripan gejala di antara berbagai jenis gangguan. Oleh karena itu, penelitian ini bertujuan untuk mengimplementasikan algoritma Support Vector Machine (SVM) dalam klasifikasi gangguan perkembangan anak, khususnya Attention Deficit Hyperactivity Disorder (ADHD), Autism Spectrum Disorder (ASD), dan Down Syndrome. Penelitian ini menggunakan pendekatan kuantitatif dengan memanfaatkan dataset yang terdiri dari 1.786 data dan sembilan atribut prediktor yang berkaitan dengan gejala perkembangan anak. Tahap prapemrosesan data dilakukan melalui penanganan nilai yang hilang menggunakan metode Most Frequent Imputation serta proses label encoding. Dataset kemudian dibagi menjadi data pelatihan dan data pengujian dengan rasio 80:20. Model SVM dikembangkan menggunakan kernel Radial Basis Function (RBF) dan dievaluasi menggunakan confusion matrix, accuracy, precision, recall, serta F1-score. Hasil eksperimen menunjukkan bahwa model SVM yang diusulkan mampu mencapai tingkat akurasi sebesar 98,88%. Laporan klasifikasi memperlihatkan kinerja yang sangat baik pada seluruh kelas dengan nilai rata-rata precision, recall, dan F1-score mencapai 0,99. Analisis confusion matrix menunjukkan hanya terdapat empat kesalahan klasifikasi dari total 358 data pengujian, sementara kelas Down Syndrome berhasil diklasifikasikan secara sempurna tanpa kesalahan. Temuan penelitian ini menunjukkan bahwa algoritma SVM sangat efektif untuk mengklasifikasikan gangguan perkembangan anak dan memiliki potensi besar dalam mendukung proses identifikasi dini serta pengambilan keputusan di bidang kesehatan anak. Model yang diusulkan dapat digunakan sebagai alat bantu pengambilan keputusan yang andal bagi tenaga kesehatan maupun orang tua dalam mendeteksi gangguan perkembangan anak berdasarkan gejala yang diamati.
Kata Kunci: Support Vector Machine (SVM), Machine Learning, Gangguan Perkembangan Anak, ADHD, Autisme, Down Syndrome, Klasifikasi.