Claim Missing Document
Check
Articles

Found 2 Documents
Search

IMPLEMENTASI METODE K-NEAREST NEIGHBOR UNTUK MENDETEKSI TINGKAT STRES PELAJAR BERDASARKAN TINGKAT PERUNDUNGAN Solihah, Siti Luthfiatin; Agusya, Kefas Zefanya; Aulia, Salman Fariz; Asfi, Marsani
Jurnal Aplikasi Bisnis dan Komputer Vol 5, No 1 (2025): Jurnal Aplikasi Bisnis dan Komputer
Publisher : Universitas Pakuan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33751/jubikom.v5i1.11608

Abstract

ABSTRAK Kelahiran seorang anak membawa harapan dan kebahagiaan bagi orang tua dan diiringi doa agar tumbuh dalam keadaan sehat jasmani serta rohani. Namun, tantangan muncul ketika anak memasuki pendidikan anak usia dini dan kemudian menghadapi risiko perundungan atau bullying di berbagai tingkatan pendidikan. Penyebabnya melibatkan faktor internal dan eksternal, seperti dorongan kekuatan diri dan pengaruh lingkungan. Oleh karena itu, pendidikan karakter sejak dini sangat penting bagi pelajar guna mengetahui sikap yang baik dan buruk di lingkungan. Pendidikan karakter berkaitan dengan sikap atau perilaku baik seseorang yang menjadi ciri dari kepribadian makhluk sosial. Pendidikan Karakter di contohkan sebagai perilaku manusia terhadap Tuhan, lingkungan sekitar, dan sesama makhluk sosial berdasarkan norma-norma yang diterapkan. Pada penelitian kali ini akan dilakukan implementasi untuk mendeteksi tingkat stres pelajar berdasarkan tingkat perundungan. Penelitian dilakukan dengan metode deskriptif dalam pendekatannya yang menggunakan pengkondisian manual pada bahasa pemrograman python dan machine learning menggunakan algoritma K-Nearest Neighbor. Metode K- Nearest Neighbor (K-NN) sebagai cara sederhana yang dilakukan pada pembelajaran mesin, menggunakan jarak Euclidean untuk mengklasifikasikan objek baru berdasarkan kedekatannya dengan k objek terdekat. Hasil dari implementasi metode K-Nearest Neighbor menunjukan keefektifannya dalam mendeteksi tingkat stres pelajar dengan dataset yang berisi 1100 data yang diberikan. Uji coba algoritma K-Nearest Neighbor (K-NN) dilakukan terhadap data stres pelajar guna mengetahui tingkat akurasi model yang digunakan. Metode ini memanfaatkan algoritma machine learning untuk melakukan prediksi berdasarkan data yang telah dilatih sebelumnya. Hal ini memberikan solusi yang dapat diandalkan dalam mendeteksi tingkat stres pelajar, terutama ketika memiliki sejumlah besar data pelatihan yang mencakup berbagai kondisi. Kata Kunci: Bullying, Euclidean, K-Nearest Neighbor, Pelajar, Pendidikan Karakter ABSTRACT The birth of a child brings hope and happiness to parents and is accompanied by prayers to grow up physically and mentally healthy. However, challenges arise when children enter early childhood education and then face the risk of bullying at various levels of education. The causes involve both internal and external factors, such as self-empowerment and environmental influences. Therefore, early character education is very important for students to know good and bad attitudes in the environment. Character education is related to a person's good attitude or behavior that characterizes the personality of a social being. Character education is exemplified as human behavior towards God, the environment, and fellow social creatures based on the norms applied. In this research, an implementation will be carried out to detect the stress level of students based on the level of bullying. The research is conducted with a descriptive method in its approach that uses manual conditioning in the python programming language and machine learning using the K-Nearest Neighbor algorithm. The K-Nearest Neighbor (K-NN) method as a simple way to do machine learning, uses Euclidean distance to classify new objects based on their proximity to the k nearest objects. The results of the implementation of the K-Nearest Neighbor method show its effectiveness in detecting student stress levels with a dataset containing 1100 given data. The K-Nearest Neighbor (K-NN) algorithm was tested on student stress data to determine the accuracy of the model used. This method utilizes machine learning algorithms to make predictions based on pre-trained data. This provides a reliable solution in detecting student stress levels, especially when having a large amount of training data covering various conditions. Keywords: Bullying, Euclidean, K-Nearest Neighbor, Student, Character Education
Pembuatan Sistem Tracer Study Menggunakan Metode Lean Software Development untuk Meningkatkan Efisiensi Proses Pengumpulan Data Aulia, Salman Fariz; Subagio, Ridho Taufiq; Supriyadi, Agung
Jurnal Ilmiah Teknik Informatika dan Komunikasi Vol. 5 No. 3 (2025): November: Jurnal Ilmiah Teknik Informatika dan Komunikasi 
Publisher : Lembaga Pengembangan Kinerja Dosen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/juitik.v5i3.1533

Abstract

Catur Insan Cendekia University (UCIC) currently utilizes manual processes involving Google Forms and Excel for tracer study implementation, resulting in inefficient alumni data collection and processing characterized by error susceptibility and fragmentation. The existing alumni portal further fails to deliver value-added resources—such as statistical tracer study analytics or structured alumni databases—impeding curriculum evaluation and evidence-based decision-making. This research develops a web-based tracer study system employing Lean Software Development (LSD) methodology to enhance data management efficiency and accuracy. The technical architecture incorporates Laravel framework (PHP), MySQL database management, and Visual Studio Code. System design utilizes UML diagrams—including use case, activity, sequence, and class diagrams—to ensure functional alignment with stakeholder requirements. Blackbox testing validation confirms optimal performance of core functionalities: user authentication, questionnaire completion, alumni data administration, and tracer study monitoring across predefined scenarios. The system successfully replaces manual workflows, mitigates data loss risks, and streamlines alumni status tracking with improved questionnaire response rates. Consequently, this solution demonstrates viability for implementing structured and integrated tracer study processes at UCIC. System design utilizes Unified Modeling Language (UML) diagrams—including use case, activity, sequence, and class diagrams—to ensure functional alignment with stakeholder requirements. These diagrams guide the development team in creating an intuitive and user-friendly interface, ensuring that all stakeholders—such as alumni, administrative staff, and faculty—are able to easily interact with the system and access relevant data. Blackbox testing validation confirms optimal performance of core functionalities: user authentication, questionnaire completion, alumni data administration, and tracer study monitoring across predefined scenarios. The system's rigorous testing process also includes stress tests to ensure it can handle large volumes of data without performance degradation.