Karim Ouda - Home

My Data Science Portfolio

A passionate freelance Data Scientist with a focus on classification and NLP problems. MSc in Data Analytics from Leeds University, built a Social Media Analytics startup from scratch. Currently advising companies on how to best use their data to optimize business decisions.

Services I offer

Portfolio


Top 10 Customer Friendly Countries based on Google Maps (Research)

Github Repository: https://github.com/karimouda/country-friendliness-index

Technologies used: R | Node.js

Other articles

Building custom visualization tools for Sterling International


Technologies used: D3, Javascript, HTML & CSS

EDA & Prediction Models for 1213BST

  • General data consultation
  • Conducting EDA for their dataset
  • Creating a sales prediction model to optimize business operations
  • Deploying the model and creating a prediction API

Technologies used: R | Python | AWS | Docker

Helping a wellbeing startup build a BI system from scratch

Hands-on support for fetching data from different sources such as Facebook Ads and Google Analytics APIs, then storing it in a centralized database for future analysis and visualization.

Technologies used: Python | MySQL | AWS | Facebook APIs

Product Manager for a Data Product @ Zalando

Overseeing the Search Ranking feature on Zalando's website

Technologies & tools used: Jira | Google Analytics

Kaggle Competition: Home Depot Product Search Relevance

Rank: 26

Predicting the relevance of search results for homedepot.com

https://www.kaggle.com/c/home-depot-product-search-relevance/leaderboard

Technologies used: R

Kaggle Competition: TalkingData Mobile User Demographics

Rank: 59

Predicting users’ demographic characteristics based on their app usage, geolocation, and mobile device properties

https://www.kaggle.com/c/talkingdata-mobile-user-demographics/leaderboard

Technologies used: R | Python

MSc Graduation Project: QuranAnalysis

Semantic Search and Intelligence System for the Quran. The system includes a smart “Search Engine”, concept exploration and visualization, 20 views for Computational Linguists such as TFIDF frequency tables, word clouds, concordance, n-grams, word similarity, PoS query and more.
https://qurananalysis.com

Github Repository: https://github.com/karimouda/qurananalysis

Technologies used: PHP | D3 | HTML | CSS | Javascript

Document Classification using WEKA (Tutorial)

One of the top WEKA tutorials on the internet. Guiding 1) How to approach a document classification problem using WEKA 2) What are the options available in WEKA to prepare your dataset for Machine Learning classification algorithms 3) Which algorithms works best for this problem 4) How to evaluate the results

Technologies used: WEKA

Founding 25trends.me - Social Media Analytics Startup

I managed to bootstrap this company as a sole entrepreneur building the whole technical stack. Conducted analysis for different purposes (clients, marketing, etc)

Technologies used: PHP | Highcharts | HTML | CSS | Javascript | Social Media APIs | Google Analytics

Tableau Portfolio

Technologies used: Tableau

Certifications


Google Analytics Individual Qualification. 2017

Let's Talk