meetups-analysis.R

library(dplyr)

ctmeet <- read.csv("meetup_groups.csv", quote = "", 
                   row.names = NULL, 
                   stringsAsFactors = FALSE)

# Adding column names
colnames(ctmeet) <- c("Area","Group","Created","Neighborhood","State","Category","Members","Who")

# Converting the Date group was created into a recognized format
ctmeet$Created_Date <- as.POSIXct(ctmeet$Created/1000, origin="1970-01-01")

# Extracting the year
ctmeet$Year <- as.numeric(format(ctmeet$Created_Date,"%Y"))

# Total number of members by Area
ctmeet %>% group_by(Area) %>% summarise(Total.Members=sum(Members))


# Analyzing number of groups per city by category and building a dataframe

cat1 <- ctmeet %>% filter(Area=="Bridgeport") %>% group_by(Category) %>% summarise(Bridgeport=length(Category))

cat2 <- ctmeet %>% filter(Area=="Hartford") %>% group_by(Category) %>% summarise(Hartford=length(Category))

cat <- join(cat2, cat1, by="Category")

cat3 <- ctmeet %>% filter(Area=="New Haven") %>% group_by(Category) %>% summarise(New.Haven=length(Category))

cat <- join(cat, cat3, by="Category")

cat4 <- ctmeet %>% filter(Area=="Stamford") %>% group_by(Category) %>% summarise(Stamford=length(Category))

cat <- join(cat, cat4, by="Category")

cat5 <- ctmeet %>% filter(Area=="Waterbury") %>% group_by(Category) %>% summarise(Waterbury=length(Category))

cat <- join(cat, cat5, by="Category")


# Turn the figures into percent and add them to the dataframe

cat$Hartford.Perc <- round(((cat$Hartford/sum(cat$Hartford, na.rm=TRUE))*100), digits=2)

cat$Bridgeport.Perc <- round(((cat$Bridgeport/sum(cat$Bridgeport, na.rm=TRUE))*100), digits=2)

cat$New.Haven.Perc <- round(((cat$New.Haven/sum(cat$New.Haven, na.rm=TRUE))*100), digits=2)

cat$Stamford.Perc <- round(((cat$Stamford/sum(cat$Stamford, na.rm=TRUE))*100), digits=2)

cat$Waterbury.Perc <- round(((cat$Waterbury/sum(cat$Waterbury, na.rm=TRUE))*100), digits=2)

# Exporting the dataframe into a CSV

write.csv(cat,"ct-categories.csv")