Digital Repository

Real Time Home Valuation Predictor

Show simple item record

dc.contributor.author Rupasinghe Arachchige, Navindu Bimsara
dc.date.accessioned 2025-06-12T06:27:31Z
dc.date.available 2025-06-12T06:27:31Z
dc.date.issued 2024
dc.identifier.citation Rupasinghe Arachchige, Navindu Bimsara (2024) Real Time Home Valuation Predictor. BSc. Dissertation, Informatics Institute of Technology en_US
dc.identifier.issn 20200891
dc.identifier.uri http://dlib.iit.ac.lk/xmlui/handle/123456789/2529
dc.description.abstract It is stated by this thesis that "Real-Time Home Valuation Predictor" is an innovative project that utilizes the power of machine learning to provide precise home valuations in a millisecond time period. This study proposal strives to tackle the intricacies under the evolving real estate market, using a novel approach that integrates real-time data analysis with state-of-the-art ensemble modeling methods to enable better predicting property values than ever before. The heart of the system is a Multi-Sourced Dataset that is diligently compiled by exploiting numerous market indicators, economic insights, and property characteristics. This diverse dataset is what fuels a complex and elaborate Voting Ensemble Model, which uses diverse techniques to increase the accuracy and dependability of the model. The ensemble method incorporates the advantageous features of the specific models and in this way, improves the predictive performance of the individual machine learning models. The model's performance is demonstrated through the usage of robust evaluation metrics, namely Mean Absolute Error (MAE) being 0.1724, Mean Squared Error (MSE) being 0.0605 and Root Mean Squared Error (RMSE) being 0.2461. The Mean Absolute Percentage Error (MAPE) shows very close to real equity prices analysts and experts expected with 0.0523 error, ascertaining that Novel Voting Ensemble Model has high level of accuracy. Also, the (R²) value of 0.9098 and the EVS of 0.9099, prove the model has a robust fit to the data. These evaluation metrics have pointed out that the model is more accurate, supplying the public with correct, instant estimations for home values. Alongside that this system involves Explainable AI (XAI) techniques, which, in turn, ensures transparency and provides end-users with an explanation of valuation outcomes, as a result, engendering trust in and understanding of the appraisal. The thesis covers the real estate market valuation and the area of machine learning through its dense and fruitful scientific work. "Real-Time Home Valuation Predictor" which is an example of efficient use of advanced algorithms along with comprehensive market study was created to meet the needs for accuracy and interpretability of property valuation tools. en_US
dc.language.iso en en_US
dc.subject Real-Time Home Valuation Predictor en_US
dc.subject Machine Learning en_US
dc.subject Ensemble Model en_US
dc.title Real Time Home Valuation Predictor en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search


Advanced Search

Browse

My Account