Welcome to my GitHub profile! I'm a data science enthusiast and a machine learning practitioner with a passion for solving real-world problems through data. I'm currently diving deep into various fields of data science and exploring innovative solutions for businesses.

I'm a data scientist focused on solving complex challenges using machine learning and artificial intelligence. I am currently working on building end-to-end machine learning pipelines and exploring the application of AI to creative domains. I have experience with a range of machine learning techniques including deep learning, NLP, and computer vision. Currently, I’m focused on applying these techniques to automate tasks in manga, like detecting dialogue bubbles and translating manga pages from Japanese to English.

• 🔭 I’m currently working on building an end-to-end ML pipeline for auto-translating manga pages from Japanese to English and YOLO-based manga bubble detection.
• 🌱 I’m learning more about Generative AI, Machine Learning Operations (MLOps), and cloud-based ML pipelines.
• 💬 Ask me about TensorFlow, YOLO, Machine Translation, Scikit-learn, Apache Airflow, Kibana, and Streamlit.
• 📫 How to reach me: LinkedIn

A comprehensive analysis of customer churn for a telecommunications company. The goal is to predict customer churn based on historical data using various machine learning models.

Generative models for text classification on the 20 Newsgroups dataset, exploring advanced techniques in natural language processing (NLP).

Analysis of traffic accident data with the aim of identifying key factors contributing to accidents and developing predictive models for accident prediction.

A project to analyze customer complaints and feedback in the telecommunications industry, leveraging sentiment analysis and topic modeling.

A study of electric vehicle (EV) charging patterns and user behavior, providing insights to improve EV infrastructure planning.

Predicting income level based on demographic and employment-related features from the UCI Census Income dataset.

Analyzing the correlation between mental health indicators (PHQ9 scores) and social media usage patterns.

A project where I used YOLOv8 to detect manga speech bubbles for automatic text extraction. The goal was to automate the extraction of dialogue from manga pages, enabling machine translation workflows.

Currently working on building an end-to-end ML pipeline that automatically translates raw manga pages from Japanese to English. This involves combining techniques from object detection, optical character recognition (OCR), and machine translation to create a seamless pipeline.

• Languages: Python, SQL
• Libraries/Frameworks: TensorFlow, Keras, Scikit-learn, Pandas, Matplotlib, Seaborn, Plotly, YOLOv8
• Databases: MySQL, PostgreSQL, MongoDB
• Cloud & Tools: AWS, Docker, Apache Airflow, Kibana, Streamlit, Elasticsearch
• Data Visualization: Matplotlib, Seaborn, Plotly, Power BI
• Other: Optical Character Recognition (OCR), NLP, Computer Vision, Generative AI

I'm currently working on:

• Building an end-to-end ML pipeline to auto-translate manga pages from Japanese to English.
• YOLO-based manga bubble detection, automating dialogue extraction from manga pages to streamline translation workflows.
• Exploring the application of Generative AI to enhance machine learning workflows, especially in creative domains like manga.

Feel free to reach out if you're interested in collaborating, discussing data science trends, or just want to connect!

• LinkedIn: handwitanto-abraham
• Email: handwitanto@gmail.com

Thank you for visiting my profile, and happy coding! ✨