Vilokana - Lightweight COVID19 Document Analysis

An increase in the scientific literature related to COVID19 makes searching for scientific information a challenging task. In this paper, we present the implementation of a semantic search engine targeted at COVID19 research articles. The algorithm uses a modified Term Frequency-Inverse Document Frequency (TFIDF) features and cosine similarity with ontology maps for semantic search. The implementation includes sentiment analysis, keyword extraction, keyword-based search, phrase extraction, textual belief indication, and text summary. The system is lightweight and can be deployed as a standalone system on mobile devices. The analysis reported in the work is based on data from 50k research articles related to COVID19 research. The tool also assists the researchers to upload their own text or documents for performing analysis. The comparisons with state of the art algorithms such as BERT and Word2Vec indicate improved performance.