Premium
Estimation of muscle pain based on Twitter data
Author(s) -
Okumura Hiroki,
Hayashi Hitoshi
Publication year - 2019
Publication title -
ieej transactions on electrical and electronic engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 30
eISSN - 1931-4981
pISSN - 1931-4973
DOI - 10.1002/tee.22889
Subject(s) - similarity (geometry) , weighting , cosine similarity , computer science , information retrieval , data mining , artificial intelligence , pattern recognition (psychology) , natural language processing , medicine , image (mathematics) , radiology
To estimate muscle pain, data from Twitter posts, or tweets, with and without symptomatic remarks were collected using the Twitter API. We performed morphological analysis, weighting by TF‐IDF, and calculation of the cosine similarity between the tweets, and then used the results to estimate muscle pain. Evaluation experiments were conducted to examine whether tweets with symptomatic remarks shared higher cosine similarity among all tweets combined. The results indicate that Twitter data could be used to characterize muscle pain symptoms. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.