Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Bulletin of Information Technology (BIT)

Analisa Prediksi Hasil Produksi Popok Bayi Metode Naïve Bayes Edy Widodo; Sifa Fauziah; Asep Arwan Sulaeman
Bulletin of Information Technology (BIT) Vol 4 No 1: Maret 2023
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bit.v4i1.504

Abstract

PT. Elleair Interantional Manufacturing Indonesia is a company engaged in the field of manufacturing baby diapers. With the increasing market demand causing an increase in the production process, what is often experienced is that there is often a lack of finish good product to meet consumer de mand due to delays in the production process. To make it easier for companies to look for factors that can increase production result, the authors coduct research with data mining using the naïve bayes method. In this study the training data and testing data were tested using the RapidMiner application with the naïve bayes algorithm where the tested data were 500 data. Testing is done by calculating the value of precision, recall, AUC dan accuracy using the RapidMiner Application and using Microsoft Excel and calculating the final probability of each class to calculate predictions of product result. With the naïve bayes method we can calculate predictions of production result based on data from the previus year as training data to anticipate shortages in production due to factors that can hider the production process. From the results of the analysis obtained factors that affect production result, namely, the number of materia used for the production of 318 data. The human error factor with the category of “No” as much as 305 data also influences because the less the occurrence of human error the production results are also high. Stop delivery factor with the category “No” as many as 299 data, with fewer cases of stop delivery, the more finish good product that can be sold
Pengelompokan Penerimaan Mahasiswa Baru Dengan Algoritma K-Means Untuk Meningkatkan Potensi Pemasaran Daniel Tambun Daniel; Sifa Fauziah; Muhtadhuddin Danny
Bulletin of Information Technology (BIT) Vol 4 No 3: September 2023
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bit.v4i3.732

Abstract

Utilization of the existing PMB dataset through the clustering method approach can be applied in analyzing the rate of acceptance of new students. The K-Medoid Cluster algorithm model that is applied has results that show a new insight, namely the grouping of new student acceptance rates based on 3 clusters, cluster 1 (C0) is a high level consisting of 49 data from 86 datasets tested and cluster 2 (C1) is a low level consisting of 11 data from 86 datasets tested and cluster 3 (C2) is a medium level consisting of 26 data from 86 datasets tested. The results of the Davies Bouldin Index or DBI value are based on the RapidMiner Studio application obtained from data testing, with a Davies-Bouldin Index evaluation value of 0.769. Keywords: Data Mining, K-Medoid Cluster, Klastrer, PMB