Claim Missing Document
Check
Articles

Found 3 Documents
Search

Optimizing the Social Assistance Recipient Model in CangkringVillage Using the Naïve Bayes Algorithm Rotika; Nana Suarna; Irfan Ali; Dendy Indriya Efendi
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.849

Abstract

Social assistance is one of the methods used by the government to help the underprivileged. Cangkring Village is a village in Cirebon Regency that has inaccurate data on recipients of social assistance or underprivileged people. The Naive Bayes algorithm is one of the most effective techniques in machine learning for classifying data, in determining the eligibility of recipients of social assistance. The method works with a probabilistic approach to analyze data efficiently and accurately, can group data based on attributes and produce high accuracy. The problem in Cangkring Village, namely the accuracy of data on recipients of social assistance, is still a problem that requires special attention. This inaccuracy not only reduces the effectiveness of social assistance programs but also creates injustice for people in need. Invalid and inappropriate data causes the distribution of social assistance to be suboptimal. The purpose of this study is to optimize the accuracy model of social security recipients using the Naive Bayes algorithm, which can help improve the accuracy in determining eligible recipients.The method used in the study is secondary data processing taken from social assistance recipient data in Cangkring Village. This process includes data preprocessing stages, training and testing data distribution, and implementation or application of the Naive Bayes algorithm to perform classification. The results of the study show that the Naive Bayes algorithm is able to increase the accuracy of the classification of social assistance recipients with an accuracy rate of 90%, compared to the conventional method used previously. This study contributes to providing a more efficient and targeted method in selecting social assistance recipients, so that it can improve the social assistance distribution system in the future. Thus, the Naive Bayes algorithm can be an effective method for data-based decision making in the context of social policy.
Improving the Voter List Clustering Model Fixed(DPT) using the K-Means Algorithm in Girinata Village Rizki Aldi; Nana Suarna2; Irfan Ali; Dendy Indriya Efendi
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.876

Abstract

Elections are one of the pillars of democracy that require accurate voter data to ensure transparency and fairness. The Permanent Voter List (DPT) is a crucial element in supporting the smooth running of elections, but there are often data validity problems such as duplicate data, voter location errors, or voter data that does not meet the requirements. This research focuses on the application of the K-Means algorithm to increase the accuracy and validity of the DPT at TPS 05, Girinata Village. The problem formulation in this research includes the accuracy level of the DPT, the effectiveness of the K-Means algorithm in identifying inaccuracies, as well as factors that influence the accuracy of voter data. This research aims to analyze the accuracy level of the DPT, evaluate the effectiveness of the K-Means algorithm in grouping data, and identify factors contributing to the validity of the DPT. The analysis results show that the K-Means algorithm succeeded in grouping voter data with good quality, with a Davies-Bouldin Index (DBI) value of 0.389, which indicates clearly defined clusters. The main factors that influence clustering are age, distance to TPS, and location (RT and TPS). This research shows that the K-Means algorithm can be used to detect inaccuracies in voter data, such as data that does not match the TPS location or age that does not meet the requirements as a voter. With these results, the K-Means algorithm makes a significant contribution to validating voter data, thereby supporting a more transparent and accountable election process.
Identify Rattan Sales Patterns Using the FP-Growth Algorithm on CV. Busaeri Rattan Robi; Nana Suarna; Irfan Ali; Dendy Indriya Efendi
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.877

Abstract

This research was conducted to recognize the pattern of purchasing rattan products at CV. Busaeri Rattan by utilizing the FP-Growth algorithm. The rattan industry is faced with the challenge of understanding consumer habits in order to improve marketing strategies. The FP-Growth algorithm was chosen for its ability to efficiently identify frequent itemset patterns without requiring a lot of memory. This research includes collecting rattan sales transaction data for one year, data preprocessing, FP-Tree structure formation, and frequent itemset analysis. The analysis was conducted using RapidMiner software with a minimum support setting of 0.005 and confidence of 0.1. The processed data was then used to find combinations of products that are often purchased together. The results revealed some significant patterns, such as the products “Mandola 3/4” and “Jawit 8/11,” which are often purchased together with a confidence level of 100%. These findings provide important insights for CV. Busaeri Rattan in increasing sales through promotional strategies such as bundling or discount offers. In addition, the FP-Growth algorithm proved to be faster and more resource-efficient than traditional methods such as Apriori. The discussion shows that the discovered purchasing patterns can help CV. Busaeri Rattan better manage stock, minimize the risk of running out of goods, and design data-driven marketing strategies. The combination of products that are often purchased together can be utilized to improve customer satisfaction as well as operational efficiency. The conclusion of this research is that the FP-Growth algorithm is an effective tool for analyzing large-scale transaction data. Further research is recommended to explore the application of this algorithm to other types of products or compare it with other data mining algorithms.