(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 13, No. 12, 2022 543 | Page www.ijacsa.thesai.org Rapid Modelling of Machine Learning in Predicting Office Rental Price Thuraiya Mohd 1 , Muhamad Harussani 2 , Suraya Masrom 3 GreensAFE (GreSFE) Research Group-Faculty of Architecture-Planning and Surveying-Department of Built Environment Studies and Technology, Universiti Teknologi MARA, Perak Branch, Seri Iskandar Campus, 32610 Perak, Malaysıa 1 Centre of Graduate Studies, Universiti Teknologi MARA, Perak Branch, Seri Iskandar Campus, 32610 Perak, Malaysıa 2 Malaysia Machine Learning and Interactive Visualization (MaLIV) Research Group Computing Sciences Study-College of Computing-Informatics and Media, Universiti Teknologi MARA, Perak Branch, Tapah Campus, 35400 Perak, Malaysia 3 Abstract—This study demonstrates the utilization of rapid machine learning modelling in an essential case of the real estate industry. Predicting office rental price is highly crucial in the real estate industry but the study of machine learning is still in its infancy. Despite the renowned advantages of machine learning, the difficulties have restricted the inexpert machine learning researchers to embark on this prominent artificial intelligence approach. This paper presents the empirical research results based on three machine learning algorithms namely Random Forest, Decision Tree and Support Vector Machine to be compared between two training approaches; split and cross- validation. AutoModel machine learning has accelarated the modelling tasks and is useful for inexperienced machine learning researchers for any domain. Based on real cases of office rental in a big city of Kuala Lumpur, Malaysia, the evaluation results indicated that Random Forest with cross-validation was the best promising algorithm with 0.9 R squared value. This research has significance for real estate domain in near future, by applying a more in-depth analysis, particularly on the relevant variables of building pricing as well as on the machine learning algorithms. Keywords—Random forest; decision tree; support vector machine; rapid prediction modelling; office rental price I. INTRODUCTION Now-a-days, real estate is becoming more digital, automated, and integrated. The fusion of industry 4.0 and digital 4.0 includes connected buildings, wearable technology, data management for buildings and infrastructure, and smart cities. The transformation of the real estate industry was improved due to the advancement in data science technologies such as analytic technologies [1]. The analytic technologies mentioned include Computational Statistics, Artificial Intelligence (AI) and Machine Learning. Machine learning is a sub-field of AI that can learn and re-learn from data exploration and inferences. Nowadays, these analytic technologies have successfully transformed the real estate industry to discover various opportunities, particularly by developing prediction applications that involve fundamental tasks that uncover hidden patterns, unknown correlations, and preferences [2]. Despite opportunities, some challenges appeared, including gaining adequate skills instantly that involves varying knowledge of AI concepts, mathematics, programming, and computer technologies. Thus, rapid software is useful to them and at the same time benefits the expert in accelerating the preliminary analytic tasks. Considering real estate markets in general, office building markets are more synchronized in terms of exposure to macro- effects and performance of the real estate within the market. The heterogeneity of the office markets makes them more complex to analyse [3], [4]. It can be challenging to understand the market, for which the property’s price might be determined on the market, but it may not always equate with the valuation of property in the market [5]. Office markets often relate to good investment opportunities since it draws much capital but with a substantial return [6]. Despite being a well-established investment industry, it has a highly complex market structure due to the lack of a central marketplace and the individuality of each building. Numerous econometric models have been proposed to predict the office market performance, especially the rental property market. These include office market econometric models [5], and the hedonic regression model [6]. Sadly, limited success was achieved in finding a reliable and consistent model to predict rental property market movements over a five-to-ten-year time frame [7]. It was expected that lacking market data can be the main problem to fault the unreliability of prediction model. Based on the preliminary statistical analysis, the collected data of office rentals has a few problems of variance insufficient, imbalance with very skewed data distribution and most of them are having low dependencies to the target data (dependent variable) to be relied by the prediction model in generating high accurate results. Acknowledging the advantages of artificial intelligence computing approach that able to learn and redevelop knowledge to self-improve the output target from the given data, the used of machine learning technique in solving issues of real estate industry has started to begin. Even with low- association dataset, machine learning with the intelligent and leaning ability, will use mathematical and heuristics projection to self-improve their performances continuously during the training stage. Despite the wider used of machine learning in various domains of problems, there is still limited work that can be found for the real estate industry. This research attempted to fill the gap by focusing on the flexible and rapid modelling machine learning approach for office rental prediction problem. This research was funded by NAPREC under grant number 100- TNCPI/GOV 16/6/2 (027/2021).