Project Overview
• Methods for data preparation and analysis, including probabilistic/Bayesian methods of data analysis, calculating correlation of features and performing feature selection.
•Unsupervised Learning and Clustering
•Supervised Learning and the problems of generalization and overfitting; Supervised learning methods including Naïve Bayes, Linear Regression, K-nearest neighbors and Decision trees
•Cutting Edge machine learning techniques: Neural Networks and Convolutional Neural networks

The Dataset
Dataset Reference: https://archive.ics.uci.edu/ml/datasets/Student+Performance
On doing analysis on this dataset, we were able to understand the factors influencing student success. The dataset consists of information collected from two Portuguese schools.

Pros:
•This dataset is recently created.
•The dataset contains 33 features so, we can explore different correlations.
•The dataset contains no missing values.
•We are familiar with the background of the dataset so it is easy to make assumptions or understand relations.
•It enables us to understand the underlying factors behind student performance and possibly take measures to better conditions.
•The features are of integer, string and Boolean types so we can carry out various analysis procedures.

Cons:
•The dataset only includes data from 2 schools in Portugal.