Twitter study, includes the Tweets between the day first COVID-19 case detected from Whuan on 31st Dec 2019, till 15th March 2020. We did twitter mining based on this dates for the study.
Source: PNGKey
Objective of this study is to do an analysis on official twitter account from the top countries' health department and WHO twitter account on COVID-19 breakout since 31st Dec 2019 in Wuhan, China. We analyzed which countries' health departments are most active in this war against the pandemic.
- Set Twitter API
- Use personal API and codes for tweet mining.
- Identify the countries official Health Department accunts on twitter
- Collect Tweets from 01/01/2020 to 15/03/2020
- Analysis on ReTweets and Active official accounts
- Analysis of frequently of words in tweets
- Combine data results and conclusion.
For the data, we turned to twitter API. Here, we targeted the important health department in countries and their official Twitter account. Twitter official accounts,
-
World Health Organization: @WHO
-
India: Ministry of Health: @MoHFW_INDIA
-
The USA: TheU.S. Department of Health and Human Services: @HHSGov
-
UK: Department ofHealth and Social Care: @DHSCgovuk
-
Australia:Australian Government Department of Health: @healthgovau
For the analysis, we collected the tweets from the respectedoffice health accounts from 31st Dec 2019 till 15th March2020.
## @WHO
tweetWHO1 = searchTwitter('from:@WHO', 2000, lang = 'en', since = '2020-01-01', until = '2020-03-15')
tweetWHO1 = do.call('rbind', lapply(tweetWHO1, as.data.frame))
View(tweetWHO1)
F0r Study purpose we focused only on ReTweets, however,
> head(tweetWHO1)
text
1 During times of stress and crisis, it is common for children to seek more attachment and be more demanding on paren… https://t.co/ttY5e8BS8L
2 We thank our 6 million followers for their trust and support to provide the world with accurate health information.… https://t.co/DGuhFme9Rq
3 RT @Refugees: "What are we doing to help refugees avoid the #coronavirus?"\n“Will people fleeing war still be able to cross borders?” \n“Coul…
4 RT @DrTedros: You do a heroic job. We know that this crisis is putting a huge burden on you and your families. We know you are stretched to…
5 RT @DrTedros: I want to send a personal and sincere thank you to every health worker around the world – especially nurses and midwives, who…
6 RT @DrTedros: Appreciated the chance to talk with @MattHancock today about #COVID19 in the <U+0001F1EC><U+0001F1E7>. @WHO is committed to working with the govern…
favorited favoriteCount replyToSN created truncated
1 FALSE 1414 WHO 2020-03-14 23:26:27 TRUE
2 FALSE 4675 <NA> 2020-03-14 23:12:56 TRUE
3 FALSE 0 <NA> 2020-03-14 23:10:57 FALSE
4 FALSE 0 <NA> 2020-03-14 23:10:33 FALSE
5 FALSE 0 <NA> 2020-03-14 23:10:30 FALSE
6 FALSE 0 <NA> 2020-03-14 23:10:12 FALSE
replyToSID id replyToUID
1 1238962852024717318 1238969771808432129 14499829
2 <NA> 1238966369477169153 <NA>
3 <NA> 1238965873932722176 <NA>
4 <NA> 1238965769611984897 <NA>
5 <NA> 1238965760455835658 <NA>
6 <NA> 1238965683158990848 <NA>
statusSource
1 <a href="http://twitter.com/download/iphone" rel="nofollow">Twitter for iPhone</a>
2 <a href="https://mobile.twitter.com" rel="nofollow">Twitter Web App</a>
3 <a href="http://twitter.com/download/iphone" rel="nofollow">Twitter for iPhone</a>
4 <a href="http://twitter.com/download/iphone" rel="nofollow">Twitter for iPhone</a>
5 <a href="http://twitter.com/download/iphone" rel="nofollow">Twitter for iPhone</a>
6 <a href="http://twitter.com/download/iphone" rel="nofollow">Twitter for iPhone</a>
screenName retweetCount isRetweet retweeted longitude latitude
1 WHO 831 FALSE FALSE NA NA
2 WHO 960 FALSE FALSE NA NA
3 WHO 1027 TRUE FALSE NA NA
4 WHO 445 TRUE FALSE NA NA
5 WHO 822 TRUE FALSE NA NA
6 WHO 132 TRUE FALSE NA NA
Similarly, we did for all the official twitter heal account from India, the US, UK and Australia.
For analysis of tweets and we focused on retweets and favoritetweets. The understanding was to showcase how effective and actives their officialhealth departments twitter official accounts are. Did they helped used calm andare those official accounts sufficient to spread awareness. Let’s check.
The ratio defines, how many tweets reTweeted by @WHO – retweeted TRUE or FALSE.
tweetWHO = tweetWHO1[, c("favoriteCount", "retweetCount", "isRetweet", 'screenName')]
# ReTweet Ratio - T/F
prop.table(table(tweetWHO$isRetweet))*100
Total_retweet_WHO = sum(tweetWHO$retweetCount)
Results indicats, only 37.4 retwetted by @WHO from the mentioned time frame. And total number of ReTweets were :Total_retweet_WHO [1] 320147
FALSE TRUE 62.58503 37.41497
Favourite tweet analysis, is for favourite tweets from @WHO. And what is the total count of Favourite tweets on @WHO tweets.
favTweetsWHO = tweetWHO[which(tweetWHO$favoriteCount != 0),]
prop.table(table(favTweetsWHO$isRetweet))*100
Total_Fav_Retweet_WHO = sum(favTweetsWHO$retweetCount)
WHO = data.frame(Total_retweet_WHO, Total_Fav_Retweet_WHO)
head(WHO)
Total ReTweets: 320147 and Total Favourite tweets: 248133
For the analysis we also created word cloud, and counts of word frequency in tweet from @WHO account.
## Check the Most Frequently used Text
textWHO = dfm(tweetWHO1$text, remove = stopwords("english"), stem = TRUE, remove_punct = TRUE)
topfeatures(textWHO, 20)
textplot_wordcloud(textWHO, min_count = 10, max_words = 100, color = c('coral', 'seagreen'),
random_order = FALSE, rotation = 0.1)
Results,
> topfeatures(textWHO, 20)
#covid19 rt @drtedro amp case countri
149 110 100 90 68 64
transmiss @who stop prevent #coronavirus communiti
61 40 39 36 35 34
spread thank respons can health support
32 28 27 26 25 22
peopl outbreak
21 19
As, we can see #covid19 and @drtredro were among the highest used words in the tweets.
NOTE: Similarly, we follow the same process to find results from respected official twitter accounts,
As we can see the following data collected from the analysis, R - code
> ORG_Tweet
[,1]
Total_retweet_WHO 320147
Total_Fav_Retweet_WHO 248133
Total_retweet_INDIA 86440
Total_Fav_Retweet_INDIA 37058
Total_retweet_AUS 1732
Total_Fav_Retweet_AUS 864
Total_retweet_USA 27603
Total_Fav_Retweet_USA 3135
Total_retweet_UK 67447
Total_Fav_Retweet_UK 36219
As we can see in the chart, WHO is leading with respect to their activeness on social medial platform Twitter. Here, we also interestingly see that the Australian health department is lagging far behind.