JDMIS: Journal of Data Mining and Information Systems

Klasifikasi Kualitas Udara di Jakarta Pada Bulan Agustus 2024 Menggunakan Algoritma C4.5

2025-01-08T07:14:38+00:00

Air is one of the important factors in human survival besides land and water. Poor air quality will have a negative impact on human health, ecosystem balance and climate change. This study aims to classify air quality in Jakarta in August 2024 using data mining techniques with the C4.5 algorithm. The data analyzed was obtained from the Satu Data Jakarta website published on February 19, 2024, including several measurement parameters, namely pm_sepuluh, pm_duakomalima, sulfur_dioxide, carbon_monoxide, ozone and nitrogen_dioxide. In its implementation, this research uses RapidMiner tools to process and analyze data. The classification results show that air quality in Jakarta during the period can be categorized into two groups, namely unhealthy and moderate, with the majority of measurements falling into the moderate category. The resulting classification model achieved an accuracy rate of 99.35%, indicating that the C4.5 algorithm is very effective in identifying and predicting air quality in Jakarta. This result shows that most of the air quality measurement data in Jakarta is still in a category that meets good air quality standards

Analisis Persepsi Publik Terhadap Pilkada Jakarta 2024 dengan Clustering dan Sentimen pada Artikel Berita

2025-01-14T02:50:27+00:00

This study analyzes the pattern of mass media coverage related to the 2024 Jakarta Gubernatorial Election using a text mining approach with the K-Means Clustering algorithm and Lexicon-based sentiment analysis. Data were obtained through web scraping from Google News RSS Feeds, resulting in 100 articles that were analyzed after undergoing preprocessing processes such as tokenizing, filtering, and stemming. The K-Means algorithm was used to cluster the articles into eight clusters based on dominant themes, such as political support, candidacy failures, and strategic issues related to the election. This algorithm works by calculating the distance between data points and centroids, which are continuously updated until an optimal cluster is achieved based on the Silhouette Score method. Sentiment analysis revealed that most articles had a neutral sentiment, reflecting media objectivity, although some clusters showed positive and negative sentiments, indicating potential bias in the coverage. These findings provide insights into the role of media in shaping public opinion and the influence of news coverage on public perceptions in the democratic process. This study is expected to enhance political literacy among the public and encourage more critical participation, while also opening opportunities for further development of analysis methods to better understand media bias

Studi Literatur Perancangan Arsitektur Data dan Aplikasi pada Perusahaan Telekomunikasi

2025-01-20T12:20:44+00:00

Enterprise Architecture (EA) serves as a blueprint for aligning business processes and IT systems to achieve organizational objectives. It provides a structured framework for integrating and optimizing various business units, enhancing overall effectiveness and maximizing outcomes. A well-defined EA is crucial for achieving business success. This research delves into the analysis of data and application modeling approaches in EA frameworks. Telecommunication companies, like other organizations, rely on EA to integrate and align their business processes and data with their corporate mission. Employing a qualitative literature review methodology, this study aims to identify the most prevalent data and application modeling approaches in EA frameworks, providing valuable insights for telecommunication companies seeking to refine their EA models in the future.

Color Space Influence on Photosynthetic Pigment Measurement Accuracy Using CNN in Color Constancy

2025-02-15T01:18:43+00:00

This study aims to design a plant pigment measurement system using digital images and deep learning, incorporating various color spaces including RGB, HSV, LAB, and YCbCr. The proposed method serves as a faster, more cost-effective, and accurate alternative to traditional methods such as spectrophotometric analysis and HPLC. Experimental results indicate that the choice of color space and inpaint preprocessing settings significantly impacts the accuracy of the CNN P3Net model. The combination of RGB+YCbCr with inpaint and RGB+LAB without inpaint yielded the lowest validation MAE values. The study also demonstrates that color constancy phenomena influence model accuracy, with color spaces that account for this phenomenon, such as RGB+YCbCr with inpaint, providing better accuracy than those that do not.

Pengembangan Aplikasi Insentif dan Komisi Salesman sebagai Strategi Peningkatan Kinerja SDM Marketing

2025-02-17T04:57:00+00:00

The performance of salespeople is greatly influenced by the incentive and commission system implemented by the company. A system that is not transparent and structured can reduce salesman motivation and productivity. This research aims to develop an application for salesman incentives and commissions as a strategy to improve marketing HR performance. The methods used in this research include requirements analysis, system design, implementation and application testing. This application is designed to be web-based to make it easier to access and increase transparency in calculating incentives and commissions. The test results show that the implementation of this application is able to increase sales force satisfaction and motivation through a fairer and real-time commission calculation system. Apart from that, the sales target tracking feature and automatic rewards contribute to improving marketing HR performance. The conclusion of this research is that the use of digital-based applications can be an effective solution in increasing transparency, trust and salesman motivation, which ultimately has a positive impact on sales productivity.

Analisis Average Waiting Time Penjadwalan CPU Menggunakan Algoritma Shortest Remaining First dan Algoritma Round Robin

2025-02-25T08:52:36+00:00

In operating systems, process scheduling is a critical aspect to determine the order of process execution by the CPU. This research compares the average waiting time (AWT) of Shortest Remaining First (SRF) algorithm and Round Robin (RR) algorithm where the problem to be solved is CPU scheduling. The purpose of this research is to get an algorithm that has a short average waiting time. The test results obtained that the SRF algorithm has a very short average waiting time with a value of 29.85 ms compared to the RR algorithm which gets an AWT result of 65.6 ms.