SEARCH-COV: A Mobile Application for Searching and Summarizing COVID-19 Research Articles

• Main Authors: Domondon, Annika Marie P., Liwag, John Joseph S., Salcedo, Lhara Ellaine L.
• Format: text
• Published: Animo Repository 2022
• Subjects: NLP; extractive summarization; Python Kivy; Python NLTK; mobile application
• Online Access: https://animorepository.dlsu.edu.ph/conf_shsrescon/2022/paper_csr/6; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1162&context=conf_shsrescon
• Institution: De La Salle University

Acquiring reliable information is necessary to combat COVID-19 effectively. However, these bits of information are contained in lengthy and highly technical research papers, which discourages the public from reading them and forces them to conveniently get information from unreliable sources that fuel misinformation. Thus, the study aimed to develop SEARCH-COV, an Android mobile application capable of searching, downloading, and automatically summarizing COVID-19 research articles. The application was coded using Python Kivy and Python Natural Language Tool Kit (NLTK). Web scraping techniques were used for developing the search functionality while extractive summarization techniques were used for the summarization functionality by implementing six different sentence scoring techniques. The first and second version of the application was evaluated using the ISO 9126 Software Quality Model for Overall Application Performance Evaluation and the International Labor Organization Qualities of a Good Summary for the Summary Quality Evaluation. After two revisions, SEARCH-COV garnered a score of 3.73 out of 5.00 for the overall performance, indicating a fair performance. On the other hand, a score of 4.12 and 4.00 out of 5.00 for the summary content and language, respectively, meaning that the summaries have good quality. This shows that SEARCH-COV has great potential to help people manage information from research articles about COVID-19 efficiently.