Matching the first two words

Hi,

I have two cell arrays A and B. A(1)= American Express Corporation B contains cells with the following string "AMERICAN EXPRESS CO" the two companies are the same but they won't match unless I ask to match the first two words (with case insensitive matching). How can I find the cells in B that matches A(1) based on the first two words only? Best, Wesso

Answer 1

[Azzi Abdelmalek]

A={'American Express Corporation','American1 Express Corporation','American2 Express Corporation'}
B={'AMERICAN EXPRESS CO','AMERICAN3 EXPRESS CO', 'AMERICAN1 EXPRESS CO' }
a=regexp(A,'\s+','split')
b=regexp(B,'\s+','split')
a1=upper(cellfun(@(x) [x{1},x{2}],a,'un',0))
b1=upper(cellfun(@(x) [x{1},x{2}],b,'un',0))

Now you can compare your companies

[Azzi Abdelmalek]

If you have

A={'abc', 'defg'}
B={'abcd' , 'efg'}

This method will not work

[Azzi Abdelmalek]

We can improve the code by adding special character like this

A={'American Express Corporation','American1 Express Corporation','American2 Express Corporation'}
B={'AMERICAN EXPRESS CO','AMERICAN3 EXPRESS CO', 'AMERICAN1 EXPRESS CO' }
a=cellfun(@strsplit,A,'un',0)
b=cellfun(@strsplit,B,'un',0)
a1=upper(cellfun(@(x) [x{1},'?' x{2}],a,'un',0))
b1=upper(cellfun(@(x) [x{1},'?' x{2}],b,'un',0))

[wesso Dadoyan]

I am facing another type of problems "Bay Banks," and "BAY BANKS" didn't match because they led to: 'BAYBANKS,' vs 'BayBanks' .How can get rid of commas and dots that come after a word?

[Guillaume]

regexp(str, '$[A-Za-z]+', 'match')

will return only the beginning of a string up to the first non-letter.

However, rather than making the string matching algorithm more and more complicated, I would rather focus my efforts in making sure the inputs were properly formatted in the first place.