I am a Lecturer in the Department of Software Engineering at Bahir Dar University, Ethiopia, with research and development experience in Artificial Intelligence, Machine Learning, Big Data Analytics, Natural Language Processing, Healthcare Informatics, Climate Informatics, and Large Language Models (LLMs).
My work focuses on developing intelligent systems that address real-world challenges in healthcare, climate resilience, education, language technologies, and sustainable development across Africa.
- 🎓 Lecturer, Department of Software Engineering, Bahir Dar University
- 🤖 AI & Machine Learning Researcher
- 📚 IEEE Author and Academic Researcher
- 🔬 Interested in interdisciplinary research combining AI,and healthcare.
- Machine Learning
- Deep Learning
- Large Language Models (LLMs)
- Retrieval-Augmented Generation (RAG)
- Explainable Artificial Intelligence (XAI)
- Federated Learning
- Reinforcement Learning
- Big Data Analytics
- Data Mining
- Predictive Analytics
- Climate Informatics
- Natural Language Processing (NLP)
- Low-Resource Language AI
- Amharic Language Processing
- Information Retrieval
- Semantic Knowledge Representation
- Image Captioning
- Medical Image Analysis
- Object Detection
- Vision-Language Models
- Medical Question Answering Systems
- Clinical Decision Support Systems
- AI for Public Health
- Biomedical NLP
- AI for Sustainable Development
- Human-Centered AI
- Responsible AI
- AI Governance and Ethics
Publisher: IEEE Xplore
This work presents a deep learning framework for generating image captions in the Amharic language using Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) architectures.
🔗 https://ieeexplore.ieee.org/document/11282808
2. Transformer-Based Deep Neural Network Model to Detect Conceptual Semantic Relations in Amharic Entities
Publisher: IEEE Xplore
This research introduces a transformer-based architecture for identifying semantic relationships among Amharic entities, contributing to knowledge extraction and language understanding for low-resource languages.
🔗 https://ieeexplore.ieee.org/document/11282590
🩺 A Production-Ready Explainable Hybrid Retrieval-Augmented Medical Question Answering System Using MedQuAD, Hybrid Dense–Sparse Retrieval, and Parameter-Efficient Fine-Tuning _Using_MedQuAD
A production-ready multilingual healthcare AI platform that enables users to ask medical questions in English and Amharic and receive evidence-based responses.
-
Explainable Medical Retrieval-Augmented Generation
-
Hybrid Dense–Sparse Medical Retrieval
-
Metadata-Aware Medical Retrieval
-
Reciprocal Rank Fusion (RRF)-based Retrieval
-
Cross-Encoder Medical Evidence Re-ranking
-
Parameter-Efficient Medical QA Fine-Tuning
-
Explainable Medical Knowledge Base Construction
-
Multi-Level Explainability
-
Confidence-Calibrated Medical QA
-
Hallucination-Aware Medical Generation -
Metadata-Guided Evidence Attribution
-
Secure Medical AI Inference Pipeline
-
Production-Ready Medical RAG Deployment
-
Comprehensive Medical QA Benchmarking
-
- FastAPI Backend Services
-
Responsive Web Interface
-
Mobile Application Support
🤗 Hugging Face Space
https://huggingface.co/spaces/samuelashu/multilingual-healthcare-rag
A production-ready multilingual healthcare AI platform that enables users to ask medical questions in English and Amharic and receive evidence-based responses.
- Retrieval-Augmented Generation (RAG)
- Fine-Tuned mT5-LoRA Language Model
- FAISS Semantic Search
- Multilingual Medical QA
- FastAPI Backend Services
- Responsive Web Interface
- Mobile Application Support
- Healthcare Knowledge Base Integration
Frontend
- Next.js
- TypeScript
- React
- Tailwind CSS
Mobile
- React Native
- TypeScript
Backend
- FastAPI
- Python
- Hugging Face Transformers
- FAISS
AI/ML
- mT5
- LoRA Fine-Tuning
- Retrieval-Augmented Generation
- Sentence Transformers
🤗 Hugging Face Space
https://huggingface.co/spaces/samuelashu/multilingual-healthcare-rag
🌐 Web Application
https://multilingual-medical-question-answe.vercel.app/
📱 Mobile Application Repository
https://github.com/samuelashu21/mobile
An AI-powered climate intelligence platform for forecasting drought risk across Ethiopian regions using satellite observations, weather datasets, machine learning models, and big data technologies.
- SPI Forecasting
- Drought Risk Classification
- Climate Data Analytics
- Regional Risk Mapping
- Machine Learning-Based Prediction
- Interactive Dashboards
- Climate Decision Support
- Python
- FastAPI
- PostgreSQL
- Redis
- Apache Spark
- Docker
- React / Next.js
- Machine Learning Models
- Python
- TypeScript
- JavaScript
- Java
- C++
- SQL\NOSQL
- TensorFlow
- PyTorch
- Scikit-Learn
- Hugging Face Transformers
- LangChain
- FAISS
- XGBoost
- LightGBM
- React
- Next.js
- React Native
- Flutter
- Node.js
- Express.js
- FastAPI
- PostgreSQL
- MongoDB
- MySQL
- Redis
- Apache Spark
- Hadoop
- Docker
- Kubernetes
- GitHub Actions
- Google Earth Engine
- GeoPandas
- Raster Analytics
- Climate Data Processing
- Large Language Models for Healthcare
- Climate Risk Prediction using Machine Learning
- Explainable AI for Decision Support Systems
- Federated Learning Security
- Low-Resource African Language Technologies
- Multilingual Question Answering Systems
- Climate Informatics and Sustainable Development
📧 Email: samuelashu21@gmail.com
🐙 GitHub: https://github.com/samuelashu21
💼 LinkedIn:
📚 Google Scholar:
"Leveraging Artificial Intelligence, Machine Learning, Big Data Analytics, and Language Technologies to create impactful solutions for healthcare, climate resilience, education, cultural heritage, and sustainable development across Africa and beyond."