Aplikasi Persediaan Barang Bangunan Cerdas Menggunakan Metode Naive Bayes Berbasis Web

Dina Lovita Sari, Feri Candra

Abstract


Application based on web can be interpreted as a program in the form of software on a particular system. Application is useful to help various activities carried out by human. CV. Cipta Sejahtera is a company engaged in the sale of various building items. The company is still using Manual Methods of Managing Data using Handwritten Accounting. The manual data collection process takes a long time, data duplication occurs by human error. Besides it reporting is often too late because it requires a long time due to having to recapitalize data on goods and data on goods expenditure. This company still does not use a system that can predict sales. The used method in this study to predict sales is Naive Bayes Methode. The Naïve Bayes method can predict goods automatically is classified as the best-selling, selling or not in demand by graphical form. This application can also order items directly on the web using Email. This application is built using 108 sales data. Where the data is divided into two parts, namely 54 training data and 54 testing data with four input variables including: Initial Stock, Stock, Number of Transactions, Transaction Items. This from application is made by using the PHP Programming Language and Mysql Database. Prediction testing in this study shows that Naive Bayes can predict sales with 98.14% accuracy and error 1.85% from 54 testing data.
Keywords: CV. Mega Cipta Sejahtera, Predict, Naive Bayes, E-mail, PHP, Database, Mysql.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.