BibTex format

author = {Gohil, SA and Vuik, S and Darzi, A},
doi = {10.2196/publichealth.5789},
journal = {JMIR Public Health and Surveillance},
title = {Sentiment analysis of health care tweets: review of the methods used},
url = {},
volume = {4},
year = {2018}

RIS format (EndNote, RefMan)

AB - Background: Twitter is a micro blogging service, where users can send and read short 140-character messages called “tweets”. There is a large amount of unstructured, free-text tweets relating to healthcare being shared on Twitter, which is becoming a popular area for healthcare research. Sentiment is a metric commonly used to investigate the positive or negative opinion within these messages. Exploring the methods used for sentiment analysis in Twitter healthcare research may allow us to better understand the options available for future research in this growing field.Objectives: The first objective was to understand which tools would be available for sentiment analysis of healthcare Twitter research, by reviewing existing studies in this area, and the methods they used. The second objective was to determine which method would work best in the healthcare settings, by anlaysing how the methods were used to answer specific healthcare questions, their production, and how their accuracy was analysed.Methods: A review of the literature was conducted pertaining to Twitter and healthcare research which used a quantitative method of sentiment analysis for the free text messages (tweets). The study compares the types of tools used in each case and examined methods for tool production, tool training and analysis of accuracy.Results: 12 papers were found studying the quantitative measurement of sentiment in the healthcare setting. More than half of these studies produced tools specifically for their research, 4 used open source tools available freely, and two used commercially available software. Five out of the twelve tools were trained using a smaller sample of the study’s final data. The sentiment method was trained against, on average, 4% of the total sample data. One of the 12 papers commented on the analysis of accuracy of the tool used. Conclusions: There are multiple methods used for sentiment analysis of tweets in the healthcare setting. These range
AU - Gohil,SA
AU - Vuik,S
AU - Darzi,A
DO - 10.2196/publichealth.5789
PY - 2018///
SN - 2369-2960
TI - Sentiment analysis of health care tweets: review of the methods used
T2 - JMIR Public Health and Surveillance
UR -
UR -
VL - 4
ER -