Python-Machine-Learning

Machine Learning Tasks using Python Programming

Predict Cancer Mortality Rates in US Counties

This project aims to predict cancer mortality rates in various US counties using machine learning techniques. The dataset includes socio-economic characteristics and other related information for specific counties.

Project Overview

The goal is to develop a regression model that can predict cancer mortality rates in "unseen" US counties. The provided dataset consists of:

• Training_data.csv: Training data with various features/predictors.
• Training_data_targets.csv: Corresponding target variables for the training set.
• Test_data_example.csv: Example test data.
• Test_data_example_targets.csv: Example test data targets.

Features

The dataset includes the following predictors/features:

• avgAnnCount: Mean number of reported cases of cancer diagnosed annually.
• avgDeathsPerYear: Mean number of deaths per year.
• incidenceRate: Mean per capita (100,000) cancer diagoses.
• medIncome: Median income.
• popEst2015: Population estimates for 2015.
• povertyPercent: Percentage of people living in poverty.
• studyPerCap: Per capita spending on studies.
• binnedInc: Income binned by 10k.
• MedianAge: Median age of residents.
• MedianAgeMale: Median age of male residents.
• MedianAgeFemale: Median age of female residents.
• GeoFips: County identifier.

Usage

1. Data Preparation: Clean and preprocess the data.
2. Model Training: Train regression models using the provided training data.
3. Model Evaluation: Evaluate the models using the example test data.
4. Inference: Use the trained model to predict cancer mortality rates on unseen test data.

Files

• Task.ipynb: Jupyter notebook containing the project code, analysis, and model development.
• Training_data.csv: Training data file.
• Training_data_targets.csv: Training targets file.
• Test_data_example.csv: Example test data file.
• Test_data_example_targets.csv: Example test targets file.

License

This project is licensed under the MIT License - see the LICENSE file for details.