- After spending 15+ years as a Software Engineer in top tier investment banks, I grew a passion for data.
- My decision to become a Data Scientist became evident, as I could easily couple my project development experience with the ability to discover patterns, classify customer behaviours and forecast future trends.
- I hold a few degrees spanning across multiple domains French Master in Banking and Finance (Merit), a MSc in Software Development (Merit), a MSc in Data Science (Distinction)
- I am a full time Innovation Manager (Full Stack Data Scientist) in a Commercial Bank's Customer Insights Analytics department.
- This Github repo contains the details of my current publications, post-grad theses, Kaggle challenges I have participated in, as well as a number of ML/AI related projects I have been working on in the past few years.
- The below sections provides a summary description of these project. You can acces the projects' documentation and code via either i) this interactive website https://rawgit.com/FredGH/ProjectPortfolio_2.0/master/index.html or ii) directly from https://github.com/FredGH/ProjectPortfolio_2.0/tree/master/projects.
- Please do not hesitate to contact me on [email protected] should you have any questions or comments.
| Category | Name | Description | Date |
|---|---|---|---|
| Conference Papers | Predicting SP500 based on its constituents and their social media derived sentiment | This work proposes, for the first time, an approach to predicting SP500 based on the closing stock prices and sentiment data of the SP500 constituents. One of the significant complexities of our framework is due to the high dimensionality of the dataset to analyse, which is based on a large number of constituents and their sentiments, and their lagging. ( Link http://research.gold.ac.uk/26368/ ) | May-2019 |
| Conference Papers | On XLE index constituents’ social media based sentiment informing the index trend and volatility prediction | This research investigates the predictive information of sentiment on the XLE index market trend and volatility, based on a novel robust machine learning approach. While we demonstrate that sentiment does not have any impact on any of the trend prediction scenarios investigated here, the sentiment’s impact on volatility predictions is significant. ( Link http://research.gold.ac.uk/24130/ ) | Aug-2018 |
| Category | Name | Description | Date |
|---|---|---|---|
| MSc Data Science Thesis | A constituent sentiment approach to stock market trend prediction | The aim is to establish the sentiment prediction power on both the XLE index's trend and volatility | Sep-2017 |
| MSc Software Development Thesis | A Model Checking Tool to detect and resolve potential Redundancies in Entity Relationship models | The aims are to prototype an enhanced version of a discovery strategy proposed by Bowers (2002) and verify how efficient and correct this solution is in discovering potential redundant relationships | Mar-2008 |
| Memoire de Maitrise Banque Finance Assurance | Credit Value-at-Risk (VaR) | Description and implementation of the VaR applied to a credit portfolio and review of the methodology limitations | Jun-1998 |
| Category | Name | Description | Date |
|---|---|---|---|
| Supervised Classification Machine Learning | Santander Customer Satisifaction Prediction | Help Santander identify dissatisfied customers early in their relationship to take proactive steps to improve customers' happiness. | Sep-2017 |
| Supervised Classification Machine Learning | Orange Customer Satisfaction Prediction | Help Orange to predict the propensity of customers to switch provider (churn)/buy new products (appetency) or upgrade (up-selling). | May-2017 |
| Category | Name | Description | Date |
|---|---|---|---|
| Supervised Learning | Stock prediction | The JPM stock prediction using supervised classification algorithm | Apr-2017 |
| Supervised Learning | Hands-on machine learning | Short exercises in machine learning Read | Jan-2017 |
| Unsupervised Learning | Creating Customer Segments | Use unsupervised techniques to see what sort of patterns exist among existing customers, and what exactly makes them different. | May-2016 |
| Supervised Learning | Student Information | Model the factors that predict how likely a student is to pass their high school final exam. | Apr-2016 |
| Supervised Learning | Predicting Boston Housing Prices | Apply basic machine learning concepts on data collected for housing prices in the Boston, Massachusetts area to predict the selling price of a new home. | Apr-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Tweeter | Sentiment Analysis | Stock Market Prediction based on Market SentimentSupervised Machine Learning on sentiment feature vectors to predict the GOOGL market trend. | May-2017 |
| NLTK | Introduction to grammar rule and NLTK | NLTK playground. | Nov-2016 |
| General | Discussion | Discussion of some of the challenges presented by Twitter and other social media as well as the use of mathematical models in Linguistics. | Oct-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Forward Chaining | Creation of an Inference Engine using unification | Implementation an inference engine for a rule-based system that operates according to the forward chaining (first-depth search) strategy and unification. | Apr-2017 |
| Search Algorithm | A Comparison of the Astar and Jump Point Search (JPS) algorithms | Implementation and performance comparison of two informed search algorithms: Astar and JSP. | Jan-2017 |
| Forward Chaining | Creation of an Inference Engine | Implementation an inference engine for a rule-based system that operates according to the forward chaining strategy (first-breath search). | Dec-2016 |
| Markov Decision Process | Self-driving smartcab | Use reinforcement learning to train a smartcab how to drive. | Jun-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Multilayer Perceptron (MLP) | MLP with Second Order Training | Build a Second Order Training MLP and show its performance profile. | Apr-2017 |
| Multilayer Perceptron (MLP) | Build a basic incremental MLP | Build a basic incremental MLP and show its performance profile. | Dec-2016 |
| Deep Learning | Build a Live Camera App | The task is to build a live camera application simulator that interprets number strings in real-world images. | Sep-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Programming | Airlines on-time performance. | A comparative study of Hadoop solutions. Implementation and comparison of a Hive and a Spark solution to answer airport delays and cancellations related questions. | Apr-2017 |
| Programming | K-means clustering with MapReduce. | Implementation of a K-means clustering algorithm into a series of Hadoop MapReduce tasks. | Feb-2017 |
| Programming | Telecom Companies Report in Spark SQL | Detailed report explaining the design/build and implementation of a reporting solution to find the top 10 customers facing frequent call drops. | Aug-2016 |
| Programming | Spark Streaming | Use Spark streaming to perform operations on streamed data. | Jul-2016 |
| Programming | Spark SQL | Perform Spark SQL operation to gather information about traded stocks. | Jul-2016 |
| Programming | Spark Rdds | Perform Rdd operations to understand customers behaviours. | Jul-2016 |
| Programming | Spark Common | Operations Use Spark to find information about a fake company employees' salary. | Jul-2016 |
| Programming | OOPS and Functional Programming in Scala | Code play ground for Object Oriented Programming in Scala. | Jul-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Programming | Flight Booking System in Cassandra | Detailed report explaining the design/build and implementation of a simplified flight booking system. | Sep-2016 |
| Notes | High Level Cassandra Concepts | Summarise some important concepts relating to the Cassandra database | Sep-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Programming | Call details record data analysis | Perform data wrangling on an Italian call details record dataset | Dec-2016 |
| Programming | Airbnb data analysis | Perform data wrangling on an Airbnb dataset | Nov-2016 |
| Programming | Telecom Companies Report in Spark SQL | Detailed report explaining the design/build and implementation of a reporting solution to find the top 10 customers facing frequent call drops. | Aug-2016 |
| Category | Name | Description | Date |
|---|---|---|---|
| Programming | The Neighborhood Map | Single-page application featuring a map of my neighborhood. | Jun-2016 |
| Programming | Frogger | Code player, enemies, and other game entities in JavaScript in object-oriented pseudo-classical style, given the art assets and game engines. | Apr-2016 |