Predictive Real Estate Analysis
This project involves leveraging Robotic Process Automation (RPA) with UiPath to perform web scraping, extracting real estate data, and subsequently utilizing data science techniques for analysis and prediction. The primary objectives include data collection, cleaning, exploration, and building a predictive model to estimate property prices based on various features such as the number of bedrooms, property type, parking spaces, and square footage.
Utilizing UiPath RPA capabilities, the project starts with web scraping to gather real estate data from online sources. UiPath is employed to navigate through web pages, extract relevant information, and store it in a structured format.
The scraped data is transferred to an Excel spreadsheet for further processing. This step involves organizing and formatting the data to prepare it for subsequent analysis.
With the data now in a usable format, the project proceeds to clean and explore the dataset. This involves handling missing values, outliers, and other inconsistencies. Exploratory data analysis (EDA) is performed to gain insights into the distribution and relationships within the dataset.
The core of the project involves building a predictive model using PyTorch, a deep learning framework. A neural network is designed and trained to predict property prices based on key features, such as the number of bedrooms, property type, parking spaces, and square footage.
In addition to the neural network, a pre-trained regression model is utilized for comparative analysis. The performance of both models is evaluated to determine their effectiveness in predicting real estate prices.
- UiPath Studio
- Python with PyTorch
- Excel for data manipulation
- Jupyter Notebook for analysis
- Clone the repository.
- Set up UiPath environment for web scraping or you can go directly to the jupyter file.
- Install required Python libraries using
pip install -r requirements.txt. - Open Jupyter Notebook for data exploration and model building.
- Execute UiPath workflow for web scraping.
- Run Jupyter Notebook for data cleaning, exploration, and model creation.
Provide insights and visualizations obtained from the analysis. Discuss the performance of the predictive model compared to the pre-trained regression model.
List potential improvements or future features to enhance the project.