X-Git-Url: https://code.communitydata.science/covid19.git/blobdiff_plain/a9f129f1d6d171346e55832887e3844b74633979..05b8025e15fe3b6f37faf5d05c1ce6850bb0b836:/transliterations/analysis/related_searches_example.R diff --git a/transliterations/analysis/related_searches_example.R b/transliterations/analysis/related_searches_example.R index 19ddfc7..e1197b6 100644 --- a/transliterations/analysis/related_searches_example.R +++ b/transliterations/analysis/related_searches_example.R @@ -3,23 +3,31 @@ ### ### Minimal example analysis file using trending search data -### Identify data source directory and file -DataDir <- ("../data/output/") -DataFile <- ("related_searches_top.csv") - ### Import and cleanup data -related.searches.top <- read.table(paste(DataDir,DataFile, - sep=""), + +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, sep=",", header=TRUE, stringsAsFactors=FALSE) +### 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) + ### 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... -top5.per.term.date$date <- asDate(top5.per.term.date$date) +top5.per.term.date$date <- as.Date(top5.per.term.date$date) ### Export write.table(top5.per.term.date,