Malva, Maisie Yunita (2024) OPTIMALISASI XGBOOST UNTUK PREDIKSI HARGA RUMAH DAN COSINE SIMILARITY UNTUK REKOMENDASI RUMAH. Project Report (Praktek Kerja Lapang dan Magang). Fakultas Ilmu Komputer.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text
21081010038-cover.pdf Download (1MB) |
|
Text
21081010038-bab1.pdf Download (275kB) |
|
Text
21081010038-bab2.pdf Restricted to Repository staff only until 8 December 2028. Download (386kB) |
|
Text
21081010038-bab3.pdf Restricted to Repository staff only until 8 December 2028. Download (699kB) |
|
Text
21081010038-bab4.pdf Restricted to Repository staff only until 8 December 2028. Download (1MB) |
|
Text
21081010038-bab5.pdf Download (250kB) |
|
Text
21081010038-daftarpustaka.pdf Download (204kB) |
|
Text
210810038-lampiran.pdf Restricted to Repository staff only until 8 December 2028. Download (1MB) |
Abstract
Housing is a primary need for people to protect themselves, rest, and store belongings. It is a fundamental element in everyone’s daily life. As times progress, home price prediction tools have become increasingly important in making purchasing decisions, as home prices vary based on building area, land area, number of bedrooms, and number of bathrooms. These four features are essential because each significantly contributes to the value and comfort of a home. The building area reflects the home’s capacity and functionality, while the land area offers future development potential and often determines the location’s prestige. The number of bedrooms indicates the home’s ability to accommodate large families or provide additional space for special needs. In contrast, the number of bathrooms directly relates to daily comfort and functionality. Considering these features helps prospective buyers make more informed decisions that align with their needs and budget. These factors make the home search process complicated and stressful for prospective buyers. Therefore, the Home Price Prediction and Recommendations System (HPPRS) was developed to help address this issue by providing accurate price predictions and suitable home recommendations based on users’ needs and preferences. This system aims to assist prospective buyers and property agents in setting competitive house prices and better assessing credit risk. This research evaluates the performance of the XGBoost model in house price prediction and recommendation systems using the Cosine Similarity method. The evaluation results show that the XGBoost model has an R-squared value of 86.63%, an MAE of 1.258381e+09%, and an MSE of 5.260574e+18%. The developed recommendation system demonstrates good performance with an MRR value of 82.58% and a Precision of 53%. These results indicate that the XGBoost model and Cosine Similarity method are effective for house price prediction and recommendations. Furthermore, this research successfully implements the model into a website for users to find house price in the South Jakarta and Tebet areas. The website is designed to provide an intuitive and informative user experience, making information supplied by HPPRS. A user-friendly interface ensures that users from various backgrounds can easily access and benefit from this service.
| Item Type: | Monograph (Project Report (Praktek Kerja Lapang dan Magang)) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||
| Depositing User: | Maisie Yunita Malva | ||||||||
| Date Deposited: | 08 Dec 2025 01:36 | ||||||||
| Last Modified: | 08 Dec 2025 01:36 | ||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/48079 |
Actions (login required)
 |
View Item |