Ensemble Learning Based Rental Apartment Price Prediction Model by Categorical Features Factoring Asif Ahmed Neloy North South University Plot, 15, Block B Kuril - NSU Rd, Dhaka 1229, Bangladesh. +880 18 6082 6446 asif.neloy@northsouth.edu H M Sadman Haque North South University Plot, 15, Block B Kuril - NSU Rd, Dhaka 1229, Bangladesh. +880 16 7722 9671 h.m.shaque03@gmail.com Md. Mahmud Ul Islam North South University Plot, 15, Block B Kuril - NSU Rd, Dhaka 1229, Bangladesh. +880 16 7344 5738 mahmud.ulislam@northsouth.e du ABSTRACT Apartment rental prices are influenced by various factors. The aim of this study is to analyze the different features of an apartment and predict the rental price of it based on multiple factors. An ensemble learning based prediction model is created to reach the goal. We have used a dataset from bProperty.com which includes the rental price and different features of apartments in the city of Dhaka, Bangladesh. The results show the accuracy and prediction of the rent of an apartment, also indicates the different types of categorical values that affect the machine learning models. Another purpose of the study is to find out the factors that signify the apartment rental price in Dhaka. To help our prediction we take on the Advance Regression Techniques (ART) and compare to different features of an apartment for establishing an acceptable model. The following algorithms are selected as the base predictors – Advance Linear Regression, Neural Network, Random Forest, Support Vector Machine (SVM) and Decision Tree Regressor. The Ensemble learning is stacked of following algorithms – Ensemble AdaBoosting Regressor, Ensemble Gradient Boosting Regressor, Ensemble XGBoost. Also, Ridge Regression, Lasso Regression, and Elastic Net Regression has been used to combine the advance regression techniques. Tree- based algorithms generate a decision tree from categorical ‘YES’ and ‘NO’ values, Ensemble methods to boosting up the learning and prediction accuracy, Support Vector Machine to extend the model for both classification and regression approach and lastly advance linear regression to predict the house price with different features values. CCS Concepts • Computing methodologies➝Ensemble methods. Keywords Ensemble learning, random forest; regularization; neural network; gradient boosting Regressor; random forest; Bangladesh apartment price prediction model. 1. INTRODUCTION One of the most widely used applications of Machine Learning is the prediction of house prices in terms of various features. An accurate and optimum rent of a house is important to homeowners, real estate, and investors. Unlike other researches, we take on the apartment price prediction in lieu of house price prediction. The rent of an apartment depends on multiple factors. Previous research models analyzed the data but most of them were simple prediction based on single forecasting model [1]. As a result, there was a huge deviation in short-term changes [2]. For addressing such issues for accurate prediction and avoid a problem, such as over-fitting, underfitting or noisy data factoring, this research suggests an ensemble learning [3]. A dataset of approximately 3505 houses entries including 19 features has been collected from bProperty.com for our research. The dataset is to be divided between an appropriate ratio for training, validation, and testing purpose. In the next section, we are going to present our model, the training process, and some computational experiments. The paper concludes with remarks. 2. RELATED WORKS There have been various extensive researches conducted on House pricing through the use of Machine Learning. Most of these works have been done in the context of developed countries. Although the house rental prices in Bangladesh are not very systematic, a pattern is present and we hope to evaluate the factors that affect the prices. There has been no substantial acceptable research on the house pricing of Bangladesh. Previous works use various methods to implement a model for this type of study [4]. One of the most popular ways to predict house pricing through machine learning is the use of Linear Regression as the model contains many features affecting the price [5]. Ensemble prediction approach using various algorithms such as GBDT (Gradient Boosting Decision Tree), XGBoost (XGBT) which has been used to predict house prices in California [6]. An approach to the use of Artificial Neural Network was used to predict the house prices in New Zealand [7]. It proved to be a daunting task as the multiple features required powerful calculations from algorithms, but the results were promising. 3. USES OF THIS MODEL 3.1 Real Estate Agents and Clients The model can be used by the real estate agents or online estate websites so that an optimal result can be shown to the clients and making renting a very easy for a particular type of apartment in Dhaka. Also, for people who are looking for a house with reasonable price and with some selected features of their own. The Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. ICMLC '19, February 22–24, 2019, Zhuhai, China © 2019 Association for Computing Machinery. ACM ISBN 978-1-4503-6600-7/19/02…$15.00 DOI: https://doi.org/10.1145/3318299.3318377 350