This portfolio is a compilation of different Data Science and Data Analysis projects I have contributed to. It includes several academic, hobby, and passion projects from which I have learned to expand my understanding of my quantitative studies further. Please contact me if any questions arise from the content shown in the documents:
- Email: ArifMeighan@hotmail.co.uk
- LinkedIn: linkedin.com/arif-meighan
Steam Game Recommendation system utilizing Collaborative Filtering
Utilized alternating least squares algorithm to model the relationship between users and products. The algorithm is capable of providing novel and intuitive game recommendations based on a user's preference for genre, publisher, and developer. It is capable of cold start recommendation where a user has no background of activity to recommend from. Groups of users with similar preferences are recommended games within their group that align with their user-product interaction tuple.
Graph Neural Network model comparison
A project based on measuring the performance of graph neural networks on the coras citation dataset. Utilizing a dataset with sparse and densely connected nodes allows for an extensive testing environment to observe how the hyperparameters, aggregation methods, and architecture of graph neural networks affect the performance of a model.
Geospatial Analysis of Crime in relation to housing prices in London
I was interested in analyzing the possible correlation between housing prices and the level of crimes. I observed that ASB (Anti-Social Behaviour) was a good indicator of housing prices but not rental prices. Utilized geospatial vector and raster structures to create chloropleth visualizations of the districts with their respective crime statistics. For a clearer copy of the code please do not hesitate to directly contact me.
- Methodologies: Machine Learning, Deep Learning, Time Series Analysis, Natural Language Processing, Statistics, Explainable AI, A/B Testing and Experimentation Design, Big Data Analytics
- Languages: Python (Pandas, NumPy, Scikit-Learn, SciPy, Keras, Matplotlib), R (Dplyr, Tidyr, Caret, Ggplot2), SQL, C++
- Tools: MySQL, Git, PySpark, Google Console, Amazon Web services (AWS)