Hanna's Portfolio

Experience

Data Scientist NLP Developer

Developed a large language model (LLM) that achieved a 60% reduction in processing time for medical entities and information extraction.

For more information

Oct 2022 - Mar 2024

Teaching Assistant - Software Engineering, Statistical Modeling, and Information Retrieval

Department of Information Systems University of Haifa

Assisted in teaching undergraduate courses in Software Engineering, Statistical Modeling, and Information Retrieval, mentoring students and grading assignments.

Oct 2022 - Aug 2024

Researech Assistant

Department of Information Systems University of Haifa

Currently engaged in research on Retrieval-Augmented Generation (RAG) to develop advanced systems specifically designed for researcher users.

June 2024

Projects

threads instagram app

NLP

Text Analysis

EDA

Sequence models

Deep Learning

Machine Learning

The project analyzes user reviews of the Threads Instagram app and develops initial predictive models, exploring different text processing techniques and machine learning models while identifying and addressing data imbalance and overfitting issues.

Hanna Bawardi
May 2023

Jigsaw Multilingual Toxic Comment Classification

NLP

Text Analysis

EDA

Sequence models

Deep Learning

Machine Learning

In this project, we developed multilingual machine learning models to classify toxicity in online conversations using English-only training data, with the goal of creating fair and effective tools to support healthier discussions across different languages.

Hanna Bawardi
November 2023

Walmart Data Sceience Project

Machine Learning

EDA

K-means

GMMs

boosting

KNN

This project examines the effect of weather on Walmart sales by combining and analyzing sales and weather data, using clustering algorithms to group weather stations, and building a machine learning model to predict sales, ultimately aiding in optimizing inventory management and business strategies.

Hanna Bawardi
Apr 2022

Game Solving Using AI Algorithms

DFS

Minimax

Pruning

Heuristic

AI (artificial intelligence)

Java

this project focus on solving the sliding puzzle game using three different search algorithms: Iterative Deepening, Breadth-First Search (BFS), and A* (A-Star). The goal of the project is to rearrange a given square of squares (3x3 or 4x4) into ascending order by switching adjacent squares vertically or horizontally.

Hanna Bawardi
May 2022

Knight's Move A Chess Inspired Game

Java

Agile

Junit

UML

Swap

Scrum

Thi project is implemetation for a chess-inspired game called Knight's Move. The game incorporates agile development and software quality assurance principles to create a fun and engaging experience. The project is developed using Java and follows the Model-View-Controller (MVC) architectural pattern.

Hanna Bawardi
October 2022

Data Science Papaer Implemetation and ML Algorithms From Scratch

Attention Is All You Need

Transformers

Language Modeling

Python

Keras

Tensorflow

NumPy

Machine Learning

Deep Learning

Papaer Implemetation

I implemented the model from the "Attention Is All You Need" paper using the NumPy and TensorFlow libraries. This implementation is from scratch, without using any pre-built Transformer or attention blocks. I also used ChatGPT to refine the language (English) and improve the clarity of the code.

Hanna Bawardi
May 2024

K-Medoids

K-Means

K-Medoids Analysis

Numpy

Python

Machine Learning

From Scratch

I implemented the model K-Medoids from scratch using only numpy (no other tools)

Hanna Bawardi
july 2024

Decision Tree And Adaboost From Scratch

Machine Learning

boosting

Decision Tree

From Scratch

Python

NumPy

I implemented a Decision Tree algorithm for both classification and regression tasks, as well as an AdaBoost classifier algorithm in Python. The aim of this project is to understand the core concepts behind these machine learning algorithms and to practice their implementation from scratch.

Hanna Bawardi
May 2022

Deep Linear Neural Network

Machine Learning

Deep Learning

DNN

From Scratch

Python

NumPy

I implemented a deep linear neural network using numpy only.

Hanna Bawardi
Jun 2023

certifications

IBM Machine Learning Professional Certificate

Machine Learning Capstone

Deep Learning

Reinforcement Learning

Supervised Machine Learning

Classification

EDA

Regression

Unsupervised Machine Learning

Python

Completed the IBM Professional Certificate in Machine Learning, gaining expertise in essential Machine Learning techniques, including Unsupervised Learning, Supervised Learning, Deep Learning, and Reinforcement Learning, with practical experience in Time Series and Survival Analysis.

Deep Learning (DeepLearning.AI | Coursera)

Structuring Machine Learning Projects

Hyperparameter Tuning

Regularization

Python

Optimization

Sequence Models

Convolutional Neural Networks

Neural Networks

completed the Deep Learning Specialization, gaining expertise in neural network architectures, including CNNs, RNNs, LSTMs, and Transformers, along with strategies like Dropout and BatchNorm. They applied these skills using Python and TensorFlow in real-world cases such as speech recognition and NLP, preparing them to advance in AI technology.

Mathematics for Machine Learning and Data Science

Linear Algebra

Calculus

Probability

Statistics

completed the Mathematics for Machine Learning and Data Science Specialization, gaining a strong foundation in core mathematics, including linear algebra, calculus, probability, and statistics, essential for advancing in AI and machine learning.

Natural Language Processing with Classification and Vector Spaces

Natural Language Processing

naïve Bayes

word vectors

locality-sensitive hashing

vector space models

machine translation

sentiment analysis

completed the course "Natural Language Processing with Classification and Vector Spaces," where I gained skills in using logistic regression, naïve Bayes, and word vectors for tasks such as sentiment analysis, analogy completion, and word translation. I also developed expertise in machine translation, locality-sensitive hashing, and vector space models.

Natural Language Processing with Probabilistic Models

Natural Language Processing

N-gram Language Models

Word2vec

dynamic programming

hidden Markov models

part-of-speech tagging

autocomplete

autocorrect

completed the course "Natural Language Processing with Probabilistic Models," where I gained skills in using dynamic programming, hidden Markov models, and word embeddings for tasks such as autocorrect, autocomplete, and part-of-speech tagging. I also developed expertise in N-gram language models and Word2vec.

Natural Language Processing with Sequence Models

Natural Language Processing

recurrent neural networks

GRU

LSTMs

Trax

named entity recognition

Natural Language Generation

completed the course "Natural Language Processing with Sequence Models," where I gained skills in using RNNs, LSTMs, GRUs, and Siamese networks for tasks such as sentiment analysis, text generation, and named entity recognition. I also developed expertise in word embedding, natural language generation, and named-entity recognition.