Prediksi Penyewaan Sepeda Menggunakan Algoritma Linear Regression Berbasis Rapidminer

  • Mohammad Agil Universitas Ibrahimy
  • Muhammad Ali Ridla Universitas Ibrahimy
Keywords: Data Mining, Linear Regression, Bike Sharing, RapidMiner, Prediksi

Abstract

The development of information technology in the current digital era has had a massive impact on various fields, including transportation. One mode of transportation widely used by the community is bike-sharing services. The high utilization of bike-sharing services makes the need for a rental quantity prediction system crucial, enabling operators to determine user demand at specific times. This study aims to predict the number of bike rentals using the Linear Regression algorithm based on data mining with the assistance of the RapidMiner application. The dataset used in this study is the Bike Sharing dataset, which consists of several attributes such as season, holiday, working day, weather, temperature, humidity, and wind speed. The research process was conducted through several stages: data collection, data preprocessing, attribute selection, splitting training and testing data, applying the Linear Regression algorithm, and evaluating the model using Performance Regression. The results show that the Linear Regression algorithm is capable of predicting the number of bike rentals quite well, based on the relationship between weather variables and the number of bike users. Temperature and weather conditions were found to be the attributes that most significantly influence the number of bike rentals. This study is expected to assist bike-sharing service operators in determining management strategies and bike provision.

Published
2026-10-28
Section
Articles