Huzaifa Abu Portfolio

Recommender Engine

A recommender engine that uses content based,collaborative and hybrid base
algorithms to suggest movies was developed. The app shows varing recommendations with the selection of each algorithm.

CLIMATE CHANGE SENTIMENT ANALYSIS

This project was used to gather the sentiments on climate change using tweets.A streamlit app was developed to allow for further analysis of text(tweets, etc.) using the generated model.

SOUTH AFRICAN LANGUAGE NLP

Using Natural Language Prosessing, 10 langauges spoken by the denizens of South Africa is classified, the oft used words in each language are also displayed by the use of word clouds.

In this project we modeled the shortfall between the energy generated by means of fossil fuels and various renewable sources - for the country of Spain. The daily shortfall, which was referred to as the target variable, was modelled as a function of various city-specific weather features such as pressure, wind speed, humidity, etc. As with all data science projects, the provided features are rarely adequate predictors of the target variable. As such, we perform feature engineering to ensure the accuracy of the model.

Analysis and Implementation of Optimization Techniques
for Facial Recognition

This is a published reseach article on techniques to improve the accuracy of face recognition algorithms in unconstrained environments. The article endeavour to find solutions to the problem of illumination,age, pose variance, occlusion and expression recognition in FR algorithms.
PCA, PSO, GA, ABC, KNN, Euclidean distance and SVM was employed during the experimentations.