JuSiTik : Jurnal Sistem dan Teknologi Informasi Komunikasi

Analisis Perilaku Siswa pada Pembelajaran Daring Menggunakan k-Means Clustering dari Log Data Moodle pada SMPN 6 Salatiga

Jeremy Evans, Yessica Nataliani — Wed, 03 Jul 2024 00:00:00 +0000

The Covid-19 pandemic in 2020 caused delays in the implementation of education in all countries, especially in Indonesia. Education must be conducted online because of the government's policy to break the chain of the spread of Covid-19. Online learning can be supported by one of the platforms, namely Moodle, which is an e-learning-based digital classroom for interaction between teachers and students. This research analyzes the behavior pattern of 215 students of grade 7 at SMPN 6 Salatiga in 2020 in the Crafts subject. Data is taken from log data Moodle. The analysis method used is k-means clustering. The research results show that the behavior of grade 7 students at SMPN 6 Salatiga in 2020 in the Crafts subject is divided into three clusters: the low activity cluster with 138 students, the medium activity cluster with 62 students, and the high activity cluster with 15 students.

Penerapan Data Mining Untuk Klasifikasi Produk Terlaris Menggunakan Algoritma Naive Bayes Pada Bengkel Motor

Alvin Julianto, Sri Andayani — Thu, 04 Jul 2024 00:00:00 +0000

Aldo Motor Workshop is one of motorcycle workshops in Palembang City, South of Sumatera.This workshop provides services for motorcycle repairs and sells motorcycle spare parts such as oil, lights, spark plugs and others. This workshop has a problem, namely that product that are in demand are often out of stock while for products that are less in demand, the stock of product has accumulated. To solve this problem, a classification data mining analysis is carried out using the naive bayes algorithm using the rapid miner application. The data used is sales data from May 2023 – October 2023. Based on the results of data processing, the accuracy results is 86.77%, precision is 85.395 and recall value is 76%. These results show that the naive bayes algorithm can be used to determine the best selling products at the Aldo Motor Workshop

TOPSIS Integration in Evaluation and Ranking of Legislative Candidates

Soetam Rizky Wicaksono — Mon, 08 Jul 2024 00:00:00 +0000

In today's dynamic electoral era, the selection of candidates for politics depends heavily on the assessment of the party or party figure on prospective legislative candidates. The TOPSIS method can provide a dynamic ranking from highest to lowest, which is one of the important ranking evaluations. Not only that, this method can also help in the selection of the best alternative from a number of alternatives, by not only considering the advantages but also disadvantages of each alternative. Thus, this research is expected to contribute to conducting a more detailed evaluation of legislative candidates dynamically and comprehensively. This research method involves several main steps: the formation of a preliminary decision matrix based on certain criteria, data normalization, weighing of the normalization matrix, determination of positive and negative ideal solutions, and calculation of distances to ideal solutions. The criteria selected include Track Record, Vision and Mission, Campaign Method, and Candidate History, with weights assigned based on importance. The Track Record and Vision and Mission were chosen because they reflect the quality and leadership ability of the candidate, while the Campaign Method and Candidate History provide an indication of the candidate's ethics and integrity. The results showed that TOPSIS was effective in ranking candidates based on predetermined criteria. Candidates with the best combination of performance across various criteria topped the rankings, while candidates with dominant negative aspects ranked lower. These results confirm the importance of holistic and balanced assessment in candidate selection, taking into account various relevant factors. The conclusion of this study shows that the application of TOPSIS in the context of candidate elections offers a transparent and accountable method to evaluate candidates, supporting the democratic process by providing objective and data-driven information to voters and decision makers.

Decision Support System for Motorcycle Credit Sales Using the Multi-Factor Evaluation Process Method at CV. Lestari Motorindo, Palembang Branch

Nanda Diane Salsabila, Nining Ariati, Dhamayanti Dhamayanti — Mon, 08 Jul 2024 00:00:00 +0000

CV. Lestari Motorindo is a company operating under PT. Astra CV. Lestari Motorindo, a subsidiary of PT. Astra Indonesia, specializes in motorcycle sales. Their credit purchase process involves gathering customer data for submission to leasing companies for verification, customer surveys, and creditworthiness assessment. Determining creditworthiness for motorcycle purchases is a complex task. Therefore, a Decision Support System (DSS) is needed to assist the marketing and admin teams in providing accurate creditworthiness recommendations.This research employs the Multi-Factor Evaluation Process (MFEP) method as the decision-making framework. MFEP is chosen due to its ability to streamline the process and enhance data validation, eliminating discrepancies between leasing companies and dealers. The MFEP-based DSS for credit applications is an accessible system for both parties, ensuring that decisions made by the DSS are valid for both leasing companies and dealers.

Text Mining Untuk Analisis Sentimen Review Film Menggunakan Algoritma Naïve Bayes

Muhammad Haidar Rifki, Yustina Retno Wahyu Utami, Paulus Harsadi — Mon, 08 Jul 2024 00:00:00 +0000

In its development, information technology brought about many changes and advances in the context of everyday life. One of the benefits of information technology to process large amounts of data is text mining. A movie is a spectacle that can be done at a relaxed time. Currently, there are many movies that can be watched via the internet or cinema. Movies that are watched on the internet are sometimes charged to watch so that potential viewers before watching a movie will read comments from users who have watched the movie. Film business and its individual reviews cannot be separated and film review sites such as IMDb is a credible source of reviews posted in public forums. Movie comments are many and varied on the IMDB website, we can see comments based on the movie title. This causes users to have difficulty analyzing other users' comments. So, this study aims to analyze the sentiment of opinions from several comments from IMDB website users and will be classified using the naïve bayes classifier method. Sentiment analysis is a classification process to understand the opinions, interactions, and emotions of a document or text. Naïve Bayes is suitable for solving multi-class prediction problems. If its assumption of the independence of features holds true, it can perform better than other models and requires much less training data. In addition to the Naïve Bayes Classifier, TF-IDF technique is also used to change the shape of the document into several words. The results obtained by applying the Naïve Bayes method are 87,2% accuracy, 92,4% recall, 80,9% precision, and 12,8% error rate.

Keywords : Classification, IMDb, Naïve Bayes, Sentiment Analysis

Sistem Pendukung Keputusan Pemilihan Media Pembelajaran online Terbaik Menggunakan Metode Smart pada Perguruan Tinggi LLDIKTI Wilayah IX

Risky Amelia Muhiddin, Muh. Maha Agung Hafid, Saharuddin Saharuddin — Wed, 31 Jul 2024 00:00:00 +0000

Along with the development of information and communication technology (ICT), in the current digital era, various kinds of learning media have emerged that can be used by higher education institutions in carrying out the teaching and learning process, including audio media, still visual media, moving visual media, audio visual media and many others. It depends on each university you want to study. SPK is designed to assist decision makers in solving both semi-structured and/or unstructured problems and focuses on presenting information that can later be used as the best choice/alternative for decision making. Online learning was first known due to the influence of the development of electronic-based learning (e-learning) introduced by the University of Illinois through a computer-based learning system. Simple Multi-Attribute Rating (SMART) is a comprehensive decision-making model that considers qualitative and quantitative aspects. The aim of this research is to make it easier for LLDIKITI IX to get information about the best online learning media, so that it can be a recommendation to universities that must be used in the online teaching and learning process