Melissa Metz

Data Analytics & Science Portfolio

Hi I'm Melissa, and I'm a business intelligence professional starting my journey in data science. In my professional experience as a product owner, technical business analyst, and data analyst, there is one recurring theme: how can businesses advance their data journey and utilize the data they have to create meaningful insights? Helping organizations understand how to solve these types of business problems is a passion of mine. My portfolio focuses on interesting projects I've recently undertaken, with a strong emphasis on business impact. Please visit my Github & LinkedIn pages (or download my Resume) by using the links below!

15 minute read

Creating An Image Search Engine Using Deep Learning

In this project we build a Deep Learning based Image Search Engine that will help customers find similar products to ones they want!...

48 minute read

Deep Learning CNN Data Science Computer Vision Python

Fruit Classification Using A Convolutional Neural Network

In this project we build & optimise a Convolutional Neural Network to classify images of fruits, with the goal of helping a groc...

13 minute read

AB Testing Hypothesis Testing Chi-Square Python

Assessing Campaign Performance Using Chi-Square Test For Independence

In this project we apply Chi-Square Test For Independence (a Hypothesis Test) to assess the performance of two types of mailers that...

15 minute read

Causal Impact Analysis Python

Quantifying Sales Uplift With Causal Impact Analysis

In this project we use Causal Impact Analysis to analyse & understand the sales uplift of customers that joined the new “Deliver...

15 minute read

Association Rule Learning Python

Understanding Alcohol Product Relationships Using Association Rule Learning

In this project we use Association Rule Learning to analyse the transactional relationships & dependencies between products in t...

12 minute read

PCA Machine Learning Python

Compressing Feature Space For Classification Using PCA

In this project we use Principal Component Analysis (PCA) to compress 100 unlabelled, sparse features into a more manageable number ...