Olivet Institute of Technology has presented its latest innovative project from the R&D center. Conducted by John Tao (MA Information Technology) as his graduate capstone, the project aims to develop a specialized search engine for Christian content on the World Wide Web.
According to Tao’s research, mainstream search engines such as Bing and Google are often not optimized for Christian keywords. For example, for the keyword “kingdom of God”, the top search result from mainstream search engines is a general definition from Britannica which does not explain how it relates to the Christian life. Tao’s idea is to build a search engine that offers more refined results and helps users set up a correct understanding of faith.
In terms of technology, the new search engine used Elasticsearch which implements text retrieval based on an inverted index mechanism. The project also used BERT (Bidirectional Encoder Representations from Transformers) for text vectorization and FAISS (Facebook AI Similarity Search) for efficient approximate nearest neighbors query. Reactjs, Flask and Cassandra were employed to build the front-end interface.
“Tao applied what he learned from the two master program tracks, namely Web Technology and Deep Learning, into this capstone project. OIT encourages students to use technology to support the propagation of the Gospel," said OIT Assistant Professor Thomas Kong.
The research project will continue throughout the quarter as Tao looks to present results at the University’s R&D Annual Showcase.