Real-Estate-Classification

Predict if two real estate listings are the same without using any location features

Data was provided through a Spanish Startup: Casafari
Developed features using a Word2Vec model to calculate cosine similarities of titles and descriptions
Applied a Histogram-based Gradient Boosting Classification Tree
81% accuracy achieved
Built an HTML webpage and Flask App to deploy the model locally simulating real time responses

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
EDA_and_Data_Cleaning.ipynb		EDA_and_Data_Cleaning.ipynb
Feature_Engineering+Modelling.ipynb		Feature_Engineering+Modelling.ipynb
Pipeline.ipynb		Pipeline.ipynb
README.md		README.md
Test_Feature_Engineering_Modelling.ipynb		Test_Feature_Engineering_Modelling.ipynb
Word2Vec_Description.ipynb		Word2Vec_Description.ipynb
Word2Vec_Title.ipynb		Word2Vec_Title.ipynb
index.html		index.html
test_flask.py		test_flask.py

Navigation Menu