2020-03-28 15:18:33 +00:00
|
|
|
### COVID-19 Digital Observatory
|
|
|
|
|
### 2020-03-28
|
|
|
|
|
###
|
|
|
|
|
### Minimal example analysis file using trending search data
|
|
|
|
|
|
|
|
|
|
### Import and cleanup data
|
2020-03-28 22:30:37 +00:00
|
|
|
|
|
|
|
|
DataURL <-
|
|
|
|
|
url("https://github.com/CommunityDataScienceCollective/COVID-19_Digital_Observatory/blob/master/transliterations/data/output/related_searches_top.csv")
|
|
|
|
|
|
|
|
|
|
related.searches.top <- read.table(DataURL,
|
2020-03-28 15:18:33 +00:00
|
|
|
sep=",", header=TRUE,
|
|
|
|
|
stringsAsFactors=FALSE)
|
|
|
|
|
|
2020-03-28 22:30:37 +00:00
|
|
|
### Alternatively, uncomment and run if working locally with full git tree
|
|
|
|
|
### Identify data source directory and file
|
|
|
|
|
## DataDir <- ("../data/output/")
|
|
|
|
|
## DataFile <- ("related_searches_top.csv")
|
|
|
|
|
|
|
|
|
|
## related.searches.top <- read.table(paste(DataDir,DataFile, sep=""),
|
|
|
|
|
## sep=",", header=TRUE,
|
|
|
|
|
## stringsAsFactors=FALSE)
|
|
|
|
|
|
2020-03-28 15:18:33 +00:00
|
|
|
### Aggregate top 5 search queries by term/day
|
|
|
|
|
top5.per.term.date <- aggregate(query ~ term + date,
|
|
|
|
|
data=related.searches.top,
|
|
|
|
|
head, 5)
|
|
|
|
|
|
|
|
|
|
## Might cleanup a bit for further analysis or visualization...
|
2020-03-28 17:01:43 +00:00
|
|
|
top5.per.term.date$date <- as.Date(top5.per.term.date$date)
|
2020-03-28 15:18:33 +00:00
|
|
|
|
|
|
|
|
### Export
|
|
|
|
|
write.table(top5.per.term.date,
|
|
|
|
|
file="output/top5_queries_per_term_per_date.csv", sep=",",
|
|
|
|
|
row.names=FALSE)
|
|
|
|
|
|
