Claim Missing Document
Check
Articles

Found 39 Documents
Search

Integration of Probabilistic Multi-Class Labeling and Adaptive K-Means Clustering with KNN Classification: Application to Weather Data Lubis, Husni; Lubis, Ihsan; Harahap, Herlina; Tommy, Tommy; Siregar, Rosyidah
Journal of Computer Science, Information Technology and Telecommunication Engineering Vol 5, No 2 (2024)
Publisher : Universitas Muhammadiyah Sumatera Utara, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30596/jcositte.v5i2.20905

Abstract

Clustering and classification technologies are pivotal in data analysis, helping to uncover hidden patterns in complex datasets. Despite their broad applications across fields such as pattern recognition, market segmentation, anomaly detection, and weather prediction, these techniques face significant limitations. Clustering methods like K-Means assume known cluster numbers and data distributions, while classification approaches such as K-Nearest Neighbors (KNN) rely heavily on the quality of labeled data. These challenges are particularly pronounced in the context of dynamic weather data, which exhibits high variability and complexity. This research addresses these limitations by integrating probabilistic multi-class labeling with an adaptive K-Means clustering approach. Probabilistic labeling allows data points to belong to multiple classes, reflecting the nuanced nature of overlapping weather conditions. Adaptive K-Means dynamically determines the optimal number of clusters, overcoming traditional constraints. By combining these methods with KNN classification, the proposed approach enhances the accuracy of weather classification. KNN leverages cluster centroids and class probabilities to provide more precise predictions. This approach provides a robust foundation for further research and optimization of adaptive methods applicable to other complex data types. Ultimately, the proposed model contributes significantly to advancing data analysis methods, particularly for dynamic and multi-class datasets like weather data.
Rancang Bangun Sistem Informasi indonesia Senti Harahap, Vira Yuanda Harahap; Ihsan Lubis; Andi Marwan Elhanafi
Jurnal Ilmu Komputer dan Sistem Informasi Vol. 2 No. 3 (2023): September 2023
Publisher : LKP Unity Academy

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70340/jirsi.v2i3.67

Abstract

CV. Medan which is located at Jalan Asia Mega Mas Blok BB No. 6, the second in the city of Yogyakarta. This company has its own product brands, namely Xidol and Eko. This company has many employees so that the performance of the company's employees must be measured properly, measuring employee performance is very important for future evaluation and planning. Evaluation of employee performance absolutely must be done to determine the achievements to be achieved by each employee. If the employee succeeds in achieving the desired target, can create a conducive atmosphere, and can satisfy all parties involved, then the employee can get a proper reward. For this reason, it is necessary to process employee appraisal data which can help make it easier for a superior to make a decision related to selecting an exemplary employee by using the Simple Addtive Weighting (SAW) method in order to be time efficient in selecting exemplary employees.
Pengenalan Website Lapak Desa Dan Cara Memasarkan Produk UMKM Secara Online Di Desa Sambi Rejo Timur Deli Serdang Sumatera Utara Lubis, Husni; Lubis, Ihsan; Halimatussa’diah; L Hutahaean, Maxi.; M Hutahaean, Marcho; Daffa , Ahmad; Artika, Yuli; Salsabila
AMMA : Jurnal Pengabdian Masyarakat Vol. 3 No. 1 (2024): AMMA : Jurnal Pengabdian Masyarakat
Publisher : CV. Multi Kreasi Media

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

Abstract

Digital marketing is one method used by business actors to introduce their products and services to the public. Apart from that, digital marketing can also be used as a transaction medium between sellers and buyers. With digital marketing, product marketing reach can be wider and reach more people. Customer and marketing costs can be reduced and made more efficient. Village Stalls is one of the service team's proposals to help business actors market  their products. In order to further promote the website, it is necessary to introduce business actors regarding the use of the website. Furthermore, the team also helps entrepreneurs or administrators with registration and provides an introduction to the features they have. From this activity, it is hoped that the marketing of business actors' products will be broader and can help increase the level of income of the people in the village.
Pelatihan Teknik Affine Cipher untuk Keamanan Data Teks Nurjamiyah, N; Lubis, Husni; Lubis, Ihsan; Irwan, Dedy; Khairani, Sumi
Jurnal TUNAS Vol 5, No 2 (2024): Edisi April
Publisher : LPPM STIKOM Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/jtunas.v5i2.113

Abstract

The development of computer technology nowadays has progressed very rapidly in line with the development of computer and telecommunication technology, making communication more open and information exchange faster, however not all developments in communication technology have had a positive and beneficial impact. One of the negative impacts of technological development is the existence of wiretapping, with the presence of encrypted tapping, the security aspect in exchanging important information. The problem of sending messages (e-mail), be its agencies, education, government, private and others, must contain confidential information, so encryption of the message is required. Encryption of information is important to support information security, because it can guarantee message security. Therefore, encryption is needed by users who have information that is guaranteed its confidentiality. One way to store data sent via email is to use cryptography. The Affine Cipher is an extension of the Caesar Cipher which multiplies the plaintext by a value and adds a shift. Creating an application by implementing cryptography using the Affine Cipher is one way of maintaining the confidentiality of information and is very important. The results obtained from this research are that the application created can change the content of existing messages and can store information that is available when sending e-mails. The process of encoding messages using the Affine Cipher algorithm is successfully used to hide secret messages into text.
Sistem Pengolahan Pemesanan Dan Pemasaran Coffee Shop Menggunakan Metode User Centered Design Mhd Alfazri; Ihsan Lubis; Boni Oktaviana Sembiring
Journal Of Informatics And Busisnes Vol. 2 No. 3 (2024): Oktober - Desember
Publisher : CV. ITTC INDONESIA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47233/jibs.v2i3.1753

Abstract

Jumlah pelanggan pada coffee shop yang banyak setiap harinya, membuat para karyawan mengalami kesulitan dalam mengelola pesanan pelanggan. Salah satunya pencatatan pesanan secara manual dengan menggunakan kertas, sehingga catatan pesanan sering mengalami miss komunikasi antara kasir dan dapur dalam pembuatan menu pesanan. Hal tersebut akan membuat pelanggan merasa tidak puas terhadap layanan dan kehilangan loyalitas pelanggan. Untuk mengatasi permasalahan tersebut dibangun sebuah sistem berbasis untuk mengelola pesanan pelanggan dan juga sebagai media promosi. Pembuatan sistem ini menggunakan metode Centered Design dan Waterfall mulai dari analisis kebutuhan, desain sistem, penulisan kode program, pengujian program, penerapan program, dan penulisan kode program dilakukan secara terstuktur. Dengan adanya sistem ini diharapkan dapat memberikan pelayanan yang lebih baik bagi pelanggan dan memberikan kemudahan bagi penjual. Hasil dari penelitian ini Sistem pengolahan pemesanan dan pemasaran coffee shop menggunakan metode user centered design telah berhasil dibuat untuk memberikan pelayanan yang lebih baik bagi pelanggan dan memberikan kemudahan bagi penjual.
Eksplorasi Data Sains: Mengubah Data Menjadi Nilai Dalam Dunia Teknologi Lubis, Husni; Lubis, Ihsan; Syahputra, Edy Rahman; Lubis, Imran; Syahputri, Nenna Irsa; Tommy, Tommy; Harahap, Herlina; Elhanafi, Andi Marwan; Siregar, Rosyidah; Khairani, Mufida
Jurnal Pengabdian Masyarakat Bangsa Vol. 2 No. 12 (2025): Februari
Publisher : Amirul Bangun Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59837/jpmba.v2i12.2058

Abstract

Penguasaan data sains menjadi keterampilan yang semakin dibutuhkan di era digital, termasuk bagi lulusan sekolah menengah kejuruan (SMK) pada bidang Teknologi Komputer dan Jaringan (TKJ) serta Rekayasa Perangkat Lunak (RPL). Namun, pembelajaran mengenai analisis data di tingkat SMK masih terbatas, sehingga siswa kurang memahami bagaimana data dapat dimanfaatkan dalam bidangnya. Pengabdian masyarakat ini bertujuan untuk mengenalkan konsep dasar data sains kepada siswa SMK Nurul Amalia Tanjung Morawa melalui pelatihan berbasis praktik. Kegiatan ini melibatkan beberapa tahap, mulai dari penyusunan materi, pelaksanaan pelatihan, hingga evaluasi. Metode pembelajaran yang digunakan adalah pendekatan teoritis dan hands-on learning melalui studi kasus yang relevan dengan dunia industri. Hasil evaluasi menunjukkan bahwa pelatihan ini meningkatkan pemahaman siswa terhadap konsep dasar data sains serta keterampilan mereka dalam melakukan analisis data sederhana. Siswa juga menunjukkan antusiasme tinggi untuk mendalami lebih lanjut bidang ini. Dari hasil kegiatan ini, disarankan agar materi data sains diintegrasikan ke dalam kurikulum SMK serta dilakukan pelatihan lanjutan untuk memperdalam keterampilan siswa dalam analisis data dan implementasi teknologinya. Selain itu, kolaborasi dengan industri dapat menjadi langkah strategis untuk memperkuat kesiapan siswa dalam menghadapi dunia kerja berbasis data.
Dari Data ke Informasi: Pengantar Sistem Informasi untuk Siswa Tommy, Tommy; Lubis, Ihsan; Lubis, Husni; Lubis, Imran; Syahputri, Nenna Irsa; Elhanafi, Andi Marwan; Siregar, Rosyidah; Khairani, Mufida; Riza, Ferdy
Jurnal Pengabdian Masyarakat Bangsa Vol. 3 No. 5 (2025): Juli
Publisher : Amirul Bangun Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59837/jpmba.v3i5.2675

Abstract

Pemahaman tentang sistem informasi merupakan kompetensi dasar yang penting bagi lulusan sekolah menengah kejuruan (SMK), khususnya pada bidang Teknik Komputer dan Jaringan (TKJ) serta Rekayasa Perangkat Lunak (RPL). Namun, materi mengenai sistem informasi secara konseptual dan aplikatif masih belum banyak dijumpai dalam kurikulum pembelajaran di tingkat SMK. Pengabdian kepada masyarakat ini bertujuan untuk mengenalkan konsep dasar sistem informasi kepada siswa SMK Nurul Amalia Tanjung Morawa melalui pendekatan edukatif dan interaktif. Kegiatan ini melibatkan beberapa tahap, mulai dari koordinasi dengan sekolah, penyusunan materi, pelaksanaan pelatihan, hingga evaluasi sederhana. Metode pembelajaran yang digunakan memadukan penyampaian teoritis, pemutaran video edukatif, diskusi kelompok, serta simulasi alur kerja sistem informasi yang relevan dengan lingkungan sekolah dan dunia kerja. Hasil kegiatan menunjukkan bahwa pelatihan ini mampu meningkatkan pemahaman siswa terhadap konsep dan penerapan sistem informasi, serta menumbuhkan minat mereka terhadap profesi di bidang teknologi informasi. Siswa juga menunjukkan antusiasme tinggi dalam sesi interaktif. Berdasarkan hasil ini, disarankan agar topik sistem informasi diperkenalkan secara lebih terstruktur dalam pembelajaran vokasional serta diikuti dengan kegiatan lanjutan yang mendorong pengembangan keterampilan aplikatif di bidang tersebut.
Penerapan Sistem HRMS Berbasis Web untuk Seleksi dan Rekomendasi Karyawan dengan Metode K – Nearest Neighbor (KNN) Anisya Avishtya Indra; Ihsan Lubis
Jurnal ilmiah Sistem Informasi dan Ilmu Komputer Vol. 5 No. 3 (2025): November: Jurnal ilmiah Sistem Informasi dan Ilmu Komputer
Publisher : Lembaga Pengembangan Kinerja Dosen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/juisik.v5i3.1521

Abstract

This research develops a website-based Human Resource Management System (HRMS) that applies the K-Nearest Neighbor (KNN) method for the selection and recommendation of permanent employees. The background to the development of this system is the challenges in the manual employee selection process, such as the large number of applicants, the difficulty of objective assessment, and the time required, which affects company productivity. The main objective of this research is to accelerate and simplify the selection process, reduce unfair assessments, and improve the accuracy of recruitment decisions. This system is designed to automatically analyze applicant data (including education level, work experience, and psychological test results) based on patterns from previous employee data. The results of this study indicate that the developed system is able to analyze applicant data and provide more accurate recommendations, significantly saving time and effort for the HR department, and producing more objective selection decisions that are in line with company needs. Thus, this system contributes to increasing efficiency, fairness, and quality in the recruitment process. In addition, the use of the K-Nearest Neighbor (KNN) method in this system provides advantages because this algorithm is able to classify data based on the level of similarity with previous data. This ensures more accurate and consistent recommendations, as they are based on historical employee patterns that have proven successful within the company. This helps HR reduce subjectivity in assessments, as selection decisions are not based solely on intuition or individual preferences, but rather on measurable, data-driven analysis. This HRMS system also has the potential for broader development, such as integration with other machine learning technologies capable of predicting long-term employee performance, or linking it to the company's performance management system.
Adaptive Categorical Dictionary Implementation for Payload Reduction in AJAX Server-side DataTables Communication Siregar, Rosyidah; Lubis, Husni; Lubis, Ihsan
Journal of Computer Science, Information Technology and Telecommunication Engineering Vol 6, No 2 (2025)
Publisher : Universitas Muhammadiyah Sumatera Utara, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30596/jcositte.v6i2.26015

Abstract

Efficient data transmission is a critical aspect of modern web applications, particularly in scenarios involving large tabular datasets rendered through server-side DataTables. This study proposes an adaptive categorical dictionary approach to reduce the payload size transmitted between the server and client. The strategy leverages the high frequency of categorical values within datasets by encoding them into shorter symbolic representations stored in a dynamically generated dictionary. The dictionary is constructed on the server during the initial request and maintained throughout the session, while the client retains a synchronized copy in memory. The research utilizes a publicly available college student dataset containing 1,545 records, focusing on columns with repetitive categorical values such as major, gender, and enrollment status. Experimental simulations were conducted under varying DataTables page lengths (10, 25, 50, and 100) to evaluate the impact of dictionary encoding on request and response payload sizes. Results demonstrate consistent payload reductions across all configurations, with significant improvements observed in larger page lengths—exceeding 12% in some cases. These findings confirm the effectiveness of the adaptive dictionary in minimizing response payloads, thereby improving communication efficiency in AJAX-based data-driven applications. The approach maintains compatibility with native PHP and JavaScript implementations and introduces minimal overhead, making it suitable for integration into existing server-side processing architectures.