Partial String Matching in R to unify text into one category

Asked Dec 18 '18 at 08:27

Active Dec 18 '18 at 08:46

Viewed 90 times

I have dataset as follow

EstablishmentName                    Freq
bahria university                    20 
bahria university islamabad          12
arid agriculture                     3
arid agriculture university          15
arid rawalpindi                      9
college of e&me, nust                20
college of e & me (nust)             15
college of eme                       30

As you can see above that Bahria University and Bahria University Islamabad are almost same, so goes for other strings. I want to unify them into one such that

Expected Output

EstablishmentName                   Freq
Bahria University                   32
Arid Agriculture                    27
College of EME                      30

I have tried the following solution but it doesn't seems to work.

     library(SnowballC)
     library(dplyr)

    mutate(df, word = wordStem(EstablishmentName)) %>%
      group_by(EstablishmentName) %>%
      summarise(total = sum(Freq))

edited Dec 18 '18 at 08:41

asked Dec 18 '18 at 08:27

Rana Usman

1,031
7
21

This is not duplicate as clearly the solutions provided there are not working. – Rana Usman Dec 18 '18 at 08:33
I have updated the post @RonakShah – Rana Usman Dec 18 '18 at 08:41
I am no expert in this, but I have used the "fuzzyjoin" packages for a similar problem. Might be useful for you aswell. – Hjalmar Dec 18 '18 at 10:44
This [post](https://stackoverflow.com/questions/11535625/similarity-scores-based-on-string-comparison-in-r-edit-distance) may help. – NRLP Dec 18 '18 at 14:53
I think you want to group by `word`, not `EstablishmentName`. – dmca Dec 18 '18 at 23:09
@dmca that doesnt changes results. – Rana Usman Dec 19 '18 at 06:16

Partial String Matching in R to unify text into one category

0 Answers0