Iyanu Akeredolu

Experience

• 

  DSNai - Data Science Nigeria

  Jun 2023 - Aug 2023

  Data Scientist

  I built a Python-based project to predict house prices using machine learning. Using the scikit-learn library, I developed a model that estimates property values based on features like location, size, and market conditions. To make it easy to use, I deployed the model with Streamlit, a tool for creating simple web apps. This project demonstrates how computer-based predictions can help in real estate, providing a useful tool for agents and buyers to make better decisions.SolutionI started by gathering housing data from various sources, ensuring it was up-to-date and accurate. After cleaning the data to remove errors and fill in missing information, I created new features to improve the model's accuracy. This involved analyzing relationships between features and house prices and introducing additional variables.For model selection, I tried different algorithms, including linear regression and Random Forest. I then fine-tuned the model's parameters using techniques like grid search and cross-validation to achieve the best results.After training the model, I evaluated its performance using metrics like Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R-squared. These metrics showed that the model was accurate and reliable. I also tested it on new data to ensure it could generalize well.To make the model accessible, I used Streamlit to build a web app where users can input data and get a predicted price. The app also includes visualizations to help users understand how the model makes predictions.Approach1. Understanding the Problem2. Data Collection3. Exploratory Data Analysis (EDA):4. Feature Engineering5. Model Selection and Training6. Hyperparameter Tuning7. Model Evaluation8. Model Deployment(Deployed the model with Streamlit, creating a user-friendly web app)9. Testing and Validation(Conducted extensive testing to ensure the model's reliability on unseen data) Show less