Title: Deep Learning Based Amharic Chatbot for FAQs in Universities

Authors: Goitom Ybrah Hailu, Shishay Welay

Published: 2024-01-26

With the intent of streamlining information dissemination in educational institutions, this paper proposes a chatbot tailored for responding to FAQs in the Amharic language. Utilizing deep learning techniques and natural language processing, the chatbot aims to alleviate the mundane task of answering repetitive questions. Here’s what you should know:

• The chatbot is designed for Amharic, a widely spoken language in Ethiopia.
• It incorporates various NLP techniques like tokenization, normalization, and stemming.
• A combination of machine learning algorithms, including deep neural networks, is employed for high accuracy.
• Integration with Facebook Messenger and deployment on Heroku server for widespread access.

This research is indicative of the strides being made in conversational AI. Not only does it cater to the specific linguistic nuances of the Amharic language, but it also exemplifies how AI can be inclusively adapted to diverse cultural contexts, enhancing accessibility and user engagement in the academic sector.