Claim Missing Document
Check
Articles

Found 7 Documents
Search

IMPLEMENTASI SISTEM ABSENSI FACE RECOGNITION BERBASIS WEB PADA BAGIAN KESEJAHTERAAN RAKYAT KABUPATEN BEKASI suprapto; Isarianto; Alhadi Saputra; Handala Simetris Harahap; Ahmad Fauzi
Jurnal SIGMA Vol 15 No 1 (2024): Juni 2024
Publisher : Teknik Informatika, Universitas Pelita Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37366/sigma.v15i1.5067

Abstract

At this time, Bagian Kesejahteraan Rakyat Kabupaten Bekasi still uses finger print attendance, the attendance process using finger print is often problematic on machines that do not detect fingers, so the attendance process is done manually by writing on the attendance form. Plus, to carry out the attendance process, employees have to queue, so it is quite a waste of time. During the Covid-19 pandemic, finger print attendance is still dangerous because of physical contact when going to the attendance process. The attendance process needs to be improved again so that its use is more flexible, safe and efficient. By utilizing face recognition technology, face recognition-based attendance is attendance that is carried out using the detection of parts of the human face. Then in the design of the face recognition-based attendance system, the researcher uses a system modeling with Undefined Modeling Language (UML) and developed with the prototype method. This research shows that with the construction of this web-based facial recognition face attendance system, Bagian Kesejahteraan Rakyat Kabupaten Bekasi can be easier and safer from the Covid-19 outbreak in carrying out attendance in every condition, then in the recapitulation of the list of employees who attend Bagian Kesejahteraan Rakyat Kabupaten Bekasi it is easier because it is already stored in the database.
Penerapan Algoritma K-Means Clustering Untuk Memetakan 4G BTS (Base Tranceiver Station) yang Mengalami Congestion di Kabupaten Bekasi Handala Simetris Harahap; Eriska Febrianto
Prosiding Sains dan Teknologi Vol. 2 No. 1 (2023): Seminar Nasional Sains dan Teknologi (SAINTEK) ke 2 - Februari 2023
Publisher : DPPM Universitas Pelita Bangsa

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

Abstract

The rapid development of wireless communication technology up to the fourth generation (4G) requires structured and systematic network data management. Datasets consisting of the number of Base Transceiver Stations (BTS), Physical Resource Block Utilization (PRB Utilization), and Downlink User Throughput (DLUT) need to be analyzed to identify areas with potential congestion and to support the planning of new BTS locations. This study aims to apply the Elbow method and K-Means Clustering to determine 4G BTS sites indicated to experience congestion in Bekasi Regency. The Elbow method was employed to identify the optimal number of clusters, resulting in K = 3. Subsequently, the K-Means algorithm was used to classify BTS based on network load levels. The results show that cluster 1 (C1), categorized as high load, consists of 153 BTS or approximately 40%, cluster 2 (C2), categorized as medium load, includes 155 BTS or about 40%, and cluster 3 (C3), categorized as low load, comprises 77 BTS or around 20%. These findings are expected to support decision-making in network optimization and 4G BTS development planning in the studied area.
Penerapan Algoritma SVM (Support Vector Machine) Untuk Prediksi Resiko Penyakit Jantung Dengan Kernel Sigmoid Handala Simetris Harahap; Safira Novianti
Prosiding Sains dan Teknologi Vol. 3 No. 1 (2024): Seminar Nasional Sains dan Teknologi (SAINTEK) ke 3 - Januari 2024
Publisher : DPPM Universitas Pelita Bangsa

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

Abstract

Heart disease, also known as coronary heart disease, occurs when blood flow to the heart muscle is reduced or blocked, causing significant damage. The objective of this study is to develop a predictive model that can estimate the risk of heart disease using the Support Vector Machine (SVM) algorithm with a sigmoid kernel, so that patients can be classified into high-risk and low-risk categories. The modeling stage is carried out to select and implement the appropriate modeling technique, determine the data mining tools to be used, and set optimal parameter values. At this stage, the training data are learned by the selected algorithm model, and the testing data are then evaluated using the developed classifier to obtain performance metrics. The results of this study indicate that the SVM method with a sigmoid kernel provides a good level of accuracy in predicting heart disease risk based on measured risk factors such as age, gender, blood pressure, cholesterol levels, and others. From the experiments conducted, the classification performed well. Using 303 data instances that were randomly sampled into 1,220 data points, the model achieved an accuracy of 0.788, a precision of 0.787, and a recall of 0.788.
Penerapan Data Mining Untuk Analisis Pola Pembelian Pelanggan Menggunakan Algoritma Apriori (Studi Kasus: Toko Jihan) Handala Simetris Harahap; Ratna Arista
Prosiding Sains dan Teknologi Vol. 3 No. 1 (2024): Seminar Nasional Sains dan Teknologi (SAINTEK) ke 3 - Januari 2024
Publisher : DPPM Universitas Pelita Bangsa

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

Abstract

Determining item combinations and item layout based on consumer purchasing tendencies is one of the solutions for Jihan Stores in developing marketing strategies so as to increase sales at the store. The algorithm that can be used to find any combination of items that are often purchased together at a time is the Apriori Algorithm, this Apriori Algorithm includes the type of rules in data mining, namely to determine associative rules between a combination of items, the results of associative rules from consumer purchasing analysis Thus, the shop owner can adjust the placement of his goods or design a marketing campaign by giving a discount on the combination of these items. Based on sales transaction data within 3 months and processed using WEKA at Jihan Stores, an analysis is carried out using an a priori algorithm with a minimum support parameter of 50% and a minimum confidence of 80%. The results of data processing with WEKA that meet the support value and the highest confidence value are that if you buy noodles, then you are likely to buy eggs.
Penerapan Metode Certainty Factor Pada Sistem Pakar Untuk Diagnosis Awal Penyakit Tuberculosis (TBC) Studi Kasus: Puskesmas Cikarang Timur Handala Simetris Harahap; Alya Shafira
Jurnal SIGMA Vol 14 No 4 (2023): Desember 2023
Publisher : Teknik Informatika, Universitas Pelita Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37366/sigma.v14i4.7317

Abstract

Tuberkulosis (TBC) merupakan penyakit infeksi yang disebabkan oleh bakteri Mycobacterium tuberculosis dan mudah menular melalui droplet di udara. Berdasarkan Global TB Report 2022, kasus TBC tertinggi di kawasan industri Cikarang terjadi pada kelompok usia produktif 25–34 tahun. Puskesmas Cikarang Timur memiliki keterbatasan tenaga ahli, sehingga diperlukan solusi untuk membantu proses diagnosis awal penyakit TBC secara lebih cepat dan efisien. Penelitian ini mengembangkan sistem pakar berbasis metode Certainty Factor yang memanfaatkan sembilan gejala utama untuk mendiagnosis kemungkinan penyakit TBC. Sistem menghasilkan output berupa persentase tingkat kemungkinan penyakit yang dialami pengguna beserta saran penanganan awal berdasarkan gejala yang dipilih. Sistem pakar yang dibangun diharapkan dapat menjadi media informasi mengenai penyakit TBC sekaligus membantu masyarakat, khususnya penderita TBC paru, dalam melakukan diagnosis awal secara praktis dengan mempertimbangkan keterbatasan waktu, jarak, dan biaya.
Pengembangan Chatbot Berbasis Retrieval Augmented Generation untuk Penyediaan Informasi Tunjangan Karyawan pada PT. XYZ Handala Simetris Harahap; Ahmad Sarif
Jurnal SIGMA Vol 16 No 1 (2025): Juni 2025
Publisher : Teknik Informatika, Universitas Pelita Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37366/sigma.v16i1.7327

Abstract

Kecerdasan buatan (AI) semakin berperan dalam industri manufaktur Indonesia seiring implementasi Strategi Nasional Kecerdasan Artifisial (Stranas KA) 2020–2045, namun keterbatasan akses informasi terkait tunjangan karyawan dalam Perjanjian Kerja Bersama (PKB) masih menjadi tantangan. Penelitian ini bertujuan mengembangkan chatbot berbasis Retrieval Augmented Generation (RAG) untuk menyediakan informasi tunjangan secara cepat dan akurat. Metode yang digunakan bersifat kuantitatif, meliputi pengumpulan dokumen PKB, pengolahan data ke dalam format terstruktur menggunakan basis data graf Neo4j, serta pengembangan sistem chatbot berbasis RAG yang terintegrasi dengan GPT-4. Pendekatan RAG mengombinasikan proses pencarian informasi dari dokumen terpercaya dengan kemampuan AI generatif untuk menghasilkan jawaban yang relevan dan kontekstual. Evaluasi sistem dilakukan melalui pengujian black-box terhadap 40 pertanyaan dengan metrik akurasi, presisi, recall, dan F1-score. Hasil pengujian menunjukkan performa yang sangat baik dengan akurasi 92%, presisi 92%, recall 100%, dan F1-score 96%, sementara pengujian langsung oleh karyawan menunjukkan tingkat keberhasilan jawaban benar sebesar 93,3%. Temuan ini menunjukkan bahwa chatbot berbasis RAG efektif dalam meningkatkan aksesibilitas informasi tunjangan karyawan serta mendukung transformasi digital di lingkungan industri.
Transformasi Digital Survei Kerusakan Kontainer: Meningkatkan Efisiensi dan Kualitas Data dengan Aplikasi MNR Berbasis Android Handala Simetris Harahap; Anggi Alfin
Prosiding Sains dan Teknologi Vol. 5 No. 1 (2026): Seminar Nasional Sains dan Teknologi (SAINTEK) ke 5 - Februari 2026
Publisher : DPPM Universitas Pelita Bangsa

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

Abstract

Transformasi digital telah menjadi kunci dalam meningkatkan efisiensi dan kualitas operasional di berbagai sektor industri, termasuk dalam survei kerusakan kontainer di bidang logistik. Penelitian ini mengkaji implementasi aplikasi MNR berbasis Android sebagai inovasi digital untuk menggantikan metode pencatatan manual dalam survei kerusakan kontainer. Dengan aplikasi tersebut, proses pengumpulan data menjadi lebih cepat, akurat, dan real-time, sehingga mendukung keputusan yang lebih tepat dan responsif. Studi ini juga membahas dampak transformasi digital terhadap peningkatan kualitas data dan efisiensi operasional melalui analisis proses kerja sebelum dan sesudah penggunaan aplikasi. Temuan menunjukkan bahwa penggunaan teknologi mobile ini mampu mengurangi kesalahan data, mempercepat alur kerja, dan meningkatkan produktivitas tim survei. Penelitian ini menegaskan pentingnya adopsi transformasi digital yang tepat guna dalam mendukung pengelolaan rantai pasok kontainer yang lebih efektif dan efisien.