Prediksi Harga Properti Dengan Machine Learning: Pendekatan Berdasarkan Spesifikasi Bangunan

Adhitya, Dimas (2023) Prediksi Harga Properti Dengan Machine Learning: Pendekatan Berdasarkan Spesifikasi Bangunan. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111840000015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (5MB) | Request a copy

Abstract

Rumah merupakan salah satu kebutuhan primer manusia. Seiring berjalannya waktu, kebutuhan akan rumah juga terus meningkat. Di sisi lain, penetapan harga rumah merupakan hal yang cukup sulit dilakukan mengingat adanya fluktuasi terhadap harga rumah. Spesifikasi bangunan dan keadaan ekonomi seperti inflasi merupakan salah satu faktor yang dianggap dapat mempengaruhi pergerakan harga rumah. Machine learning yang saat ini terus berkembang dapat dimanfaatkan untuk membuat sistem prediksi maupun menganalisis hubungan antar variabel. Pada penelitian ini dipelajari bagaimana cara memanfaatkan machine learning untuk membuat sistem prediksi harga rumah dan melihat pengaruh inflasi terhadap pergerakan harga rumah. Data yang digunakan adalah data penjualan perumahan di Ames, Iowa, Amerika Serikat dari tahun 2006 hingga 2010. Penelitian memanfaatkan beberapa algoritma machine learning yaitu Ridge Regression, LASSO, Elastic Net, Support Vector Regression, dan LightGBM untuk membandingkan akurasi dari setiap algoritma dengan berfokus pada proses feature engineering dan hyperparameter tuning. Hasil penelitian menunjukkan bahwa LightGBM mampu mengungguli algoritma lain dalam memprediksi harga rumah, di mana proses feature engineering dan hyperparameter tuning menjadi proses yang penting untuk diperhatikan agar model prediksi yang dihasilkan dapat menjadi lebih optimal. Selain itu, ditemukan juga bahwa tidak terdapat bukti yang menunjukkan pengaruh signifikan inflasi terhadap pergerakan harga rumah
====================================================================================================================================
House is one of the primary human needs. As time goes by, the need for houses also continues to increase. On the other hand, determining the price of a house is quite difficult to do considering the fluctuations in house prices. Building specifications and economic conditions such as inflation are among the factors that are considered to affect house price movements. Machine learning, which is currently developing, can be used to create a prediction system and analyze the relationship between variables. This research will study how to utilize machine learning to create a house price prediction system and see the effect of inflation on house price movements. The data used is housing sales data in Ames, Iowa, United States from 2006 to 2010. The research utilizes several machine learning algorithms namely Ridge Regression, LASSO, Elastic Net, Support Vector Regression, and LightGBM to compare the accuracy of each algorithm by focusing on the feature engineering process and hyperparameter tuning. The results show that LightGBM is able to outperform other algorithms in predicting house prices, where the feature engineering and hyperparameter tuning processes are substantial in improving the model accuracy so that the resulting prediction model can be more optimal. In addition, the study also found that there is no proof showing a significant effect of inflation on house price movements

Item Type:	Thesis (Other)
Additional Information:	RSIf 006.31 Adh p-1 2023
Uncontrolled Keywords:	Properti, Rumah, Inflasi, Machine Learning, Ridge Regression, LASSO, Elastic Net, Support Vector Regression, LightGBM, Property, House, Inflation
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Dimas Adhitya
Date Deposited:	03 Feb 2023 03:10
Last Modified:	24 Aug 2023 08:02
URI:	http://repository.its.ac.id/id/eprint/96135

Actions (login required)

View Item