발간년도 : [2021]
| 논문정보 |
|
| 논문명(한글) |
[Vol.16, No.2] Semantic-based Word Embedding Measure for COVID-19 Healthcare on Big Data Environment |
|
| 논문투고자 |
Jong-Yun Kim, Jinhong Kim |
|
| 논문내용 |
In late 2019, the world became aware of a previously unknown human pathogen, a virus that was soon named SARS-COV-2. The disease caused by the virus, called COVID-19, spread to nearly every country on Earth in less than a year, with catastrophic consequences: by midNovember 2020 at least 53 million people worldwide had contracted the virus, over 1.2 million people had died from COVID-19, and the economies in many countries were in serious straits. The world was not prepared for the pandemic, yet it was not a surprise. Global health practitioners, national governments, the World Health Organization, scholars, and pundits all had been saying for years that a global pandemic of influenza or a related disease was imminent. Accordingly, COVID-19 is an ongoing pandemic, has caused millions of people try to retrieve relevant information regarding the disease. In order to discourse the authoritative information about the disease, we propose a semantic-based Korea QAS (Query Answering System). This is an automatic domain specific knowledge system to deliver reliable answers for COVID-19 related queries in Korea language. In addition, Natural Language Processing techniques are used for processing Korea queries and documents. Word embedding method CBOW is used for document conversion and semantic modeling. Then the cosine similarity is used to rank and retrieve the corresponding results for the queries. The experiment is conducted with our own Korea Query-answering data set created from publically available, authoritative COVID-19 related information, and the accuracy of the system is evaluated 0.72% to 0.76%. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
