Qualified finance and sales management professional, with experience in Data Analysis. Passionate about gathering, analyzing, and manipulating data to meet deliverables. Able to learn new technologies, software and computers skills. Capable of utilizing my bilingual skills (English / Arabic) to create reports, provide data, and present in both languages.
I enjoy being challenged and engaging with projects that require me to work outside my comfort and knowledge set, as continuing to learn new languages and development techniques are important to me. My technical expertise is in Mac's IOS platform. Confident in a number of scripting/programming languages (including Python, JavaScript, and SQL), Markup languages (including HTML, CSS), Databases (including MySQL, MongoDB), and Frameworks (including Flask, BeautifulSoup). I also have a great knowledge of analytical, communicational, and visualization libraries (including NumPy, Pandas, SQLAlchemy, Matplotlib, Seaborn).
Working knowledge in Machine Learning and good understanding in analyzing big datasets and making prediction using (Spark, Hadoop, TensoFlow, CNN, SciKit, MapReduce ).
This website presents some selected projects I worked on, to find the full list of projects please visit my GitHub_Portfolio here.
We chose this data set for the wealth of info if offered: over 7K records of academic institutions, each record comprised of about 1.2K columns. This mountain of data was condensed/compressed into a kernel of 989 institutions with 70 columns, using a process of pruning empty and unneeded columns followed by removal of records containing privacy-suppressed fields.
We then applied supervised and unsupervised Machine Learning techniques to explore the following: 1. What are the prominent features that determine earning potential for graduating students? 2. Is there a set of features (eg major) that can be used to predict the earning potential for graduating students? 3. What, if any, natural groupings can be found in the set of features explored the above questions?
Decision Tree and PCA were used for feature selection; multi-variate regression analysis was trained on the top ten dominant features driving post graduate income; cluster analysis was applied to identify natural grouping among the top three features used in regression analysis. Findings and observations are detailed in subsequent tabs.