A collection of Machine Learning examples using python - This is a work in progress and is being updated on a regular basis - Last updated on 1st Nov 2018.

For this collection of examples, I have chosen to use local Irish datasets for analysis whenever possible. I have tried not to use the standard classic teaching examples as these have been very well documented online to date - and hence easily replicated.

I have created this repository so that I can demonstrate my abilities in the areas of Machine Learning, Charting, Data Analysis and manipulation using Python..

Posted iPython Notebooks

Linear Regression - Height & Weight	Predicting a persons probable weight based on their height. Using a training data set to create a simple linear regression model Jupiter Notebook
A/B Testing, T-Scores and P-Values - comparing two data sets for experimental purposes in python	Using sample website data we are examining the affects of a new page button design on the flow of traffic on the website. The unit of measurement is determined by clicks on the new button and the old button. The traffic would be recorded as two data sets, the traffic using the old button (control group) and the new traffic using the new button (test group).This data is then analysed to determine the new button effectiveness using the T-values and P-values of the analysis. Jupiter Notebook
Decision Trees and Random forests: Will it rain in Dun Laoghaire?	A Python based simple Decision Trees and Random forests project that examines historic weather data to predict using a decision trees method if it will rain on a given data based on a set of variables (temp, humidity, etc...) Jupiter Notebook
Logistic Regression - Looking at Leaving cert points and collage admission	A simple logistic regression in Python that predicts a students chances admission to college based on leaving cert points. Jupiter Notebook
Creating a Recommender system based upon Amazon review data	Using Amazon customer review data for musical instruments I created a recommender system that will suggest similar products that other users have rated highly along side the product that is being purchased. This wil then be used to create a 'We think you might like this!' targeted advertisement to the customer. Jupiter Notebook
K Means Clustering with Online retail data - a practical example	In this notebook we create a labelling system for each customer that allow each customer to be grouped into groups based on a particular measure. Jupiter Notebook
KNN - K Nearest Neighbours with Online retail data	In this notebook we create a labelling system for each customer which allows each customer to be grouped into groups based on the location of a groups center. Jupiter Notebook
Nasdaq Stock Share Price Analysis - Google, Netflix, Amazon, Facebook, Tesla Motors and Twitter	In this notebook we look at plotting the share values of some well known US stocks. Jupiter Notebook
Naive Bayes - Sentiment Analysis of film reviews using the NTLK package	NLTK is a powerful Python package that provides a set of diverse natural languages algorithms. It is free, opensource, easy to use, large community, and well documented. NLTK consists of the most common algorithms such as tokenizing, part-of-speech tagging, stemming, sentiment analysis, topic segmentation, and named entity recognition.The process of breaking down a text paragraph into smaller chunks such as words or sentence is called Tokenization. Token is a single entity that is building blocks for sentence or paragraph. In this project we analysis the 'mood' of film reviews to determine if a user likes or dislikes the film. Jupiter Notebook
Classification of the MNIST dataset using Scikit-Learn and Python	The MNIST dataset is a set of images of 70,0000 handwritten black and white digits of 0-9. Using Scikit Learn we use the package to recognize each small image and the number contained in the image. Jupiter Notebook