risiriccardo / finding_donors_for_charity

Finding Donors for Charity - project from Udacity's "Intro to Machine Learning with Pytorch" Nanodegree Program

Finding Donors for Charity Using Machine Learning

Project Description

This project was developed in the context of Udacity's Intro to Machine Learning with Pytorch Nanodegree.

The goal of this project is to developed a ML model (using supervised learning techniques) which is capable of predicting whether a person is likely to be a donor or not. This could help charity organizations in optmizing their marketing efforts, targeting only those who are more likely to donate. The target variable of this model is the person's income (<=50K or >50K). The key assumption being that those with a lower income are going to be less prone to donating.

As a first step, an analysis is conducted to determine which factors are the key-indicators of a person being more likely to earn a >50K salary. Then, three different supervised learning algorithms (Random Forest, Support Vector Machines, and Adaboost) are trained and evaluated, so that the best one can be used by the charity organization.

Install

This project requires Python >=2.7 and the following Python libraries installed:

• NumPy
• Pandas
• matplotlib
• scikit-learn
• IPython
• Jupyter Notebook

Code

The main code for this project is located in the finding_donors.ipynb notebook file. Additional supporting code for visualizing the necessary graphs can be found in visuals.py.

The Report.html file contains a snapshot of the main code in the jupyter notebook with all code cells executed.

Run

In the terminal, navigate to the top-level project directory finding_donors_for_charity/ and run:

jupyter notebook finding_donors.ipynb

Data

The modified census dataset consists of approximately 32,000 data points, with each datapoint having 13 features. This dataset is a modified version of the dataset published in the paper "Scaling Up the Accuracy of Naive-Bayes Classifiers: a Decision-Tree Hybrid", by Ron Kohavi. You may find this paper online, with the original dataset hosted on UCI.

Features

• age: Age
• workclass: Working Class (Private, Self-emp-not-inc, Self-emp-inc, Federal-gov, Local-gov, State-gov, Without-pay, Never-worked)
• education_level: Level of Education (Bachelors, Some-college, 11th, HS-grad, Prof-school, Assoc-acdm, Assoc-voc, 9th, 7th-8th, 12th, Masters, 1st-4th, 10th, Doctorate, 5th-6th, Preschool)
• education-num: Number of educational years completed
• marital-status: Marital status (Married-civ-spouse, Divorced, Never-married, Separated, Widowed, Married-spouse-absent, Married-AF-spouse)
• occupation: Work Occupation (Tech-support, Craft-repair, Other-service, Sales, Exec-managerial, Prof-specialty, Handlers-cleaners, Machine-op-inspct, Adm-clerical, Farming-fishing, Transport-moving, Priv-house-serv, Protective-serv, Armed-Forces)
• relationship: Relationship Status (Wife, Own-child, Husband, Not-in-family, Other-relative, Unmarried)
• race: Race (White, Asian-Pac-Islander, Amer-Indian-Eskimo, Other, Black)
• sex: Sex (Female, Male)
• capital-gain: Monetary Capital Gains
• capital-loss: Monetary Capital Losses
• hours-per-week: Average Hours Per Week Worked
• native-country: Native Country (United-States, Cambodia, England, Puerto-Rico, Canada, Germany, Outlying-US(Guam-USVI-etc), India, Japan, Greece, South, China, Cuba, Iran, Honduras, Philippines, Italy, Poland, Jamaica, Vietnam, Mexico, Portugal, Ireland, France, Dominican-Republic, Laos, Ecuador, Taiwan, Haiti, Columbia, Hungary, Guatemala, Nicaragua, Scotland, Thailand, Yugoslavia, El-Salvador, Trinadad&Tobago, Peru, Hong, Holand-Netherlands)

Target Variable

• income: Income Class (<=50K, >50K)