Track changes in a cell array with strings

I have a cell array with 5 columns and about 5000 rows with string elements. E.g.:

1997 Charles House Materials Chemicals

The years go from 1997 to 2013 and you find repetitions while the years pass by. What I would like to get is a new cell array with the cases in which the combination of the second and third column change. E.g. original cell array:

1997 Charles House Materials Chemicals %initial

1997 Rita Office Financial Bank %initial

1998 Rita Office Financial Bank %no change

1999 Charles House Materials Chemicals %no change

2000 Charles Office Materials Chemicals %change in the 2nd column

2001 Charles Office Materials Chemicals %no change

2003 Rita Star Financial Bank %change in the 2nd column

2005 Charles Castle Materials Chemicals %change in the 2nd column

2010 Rita Moon Financial Bank %change in the 2nd column

I would like for my new array to give me the first/original row and the cases in which you observe a change. E.g. output:

1997 Charles House Materials Chemicals

1997 Rita Office Financial Bank

2000 Charles Office Materials Chemicals

2003 Rita Star Financial Bank

2005 Charles Castle Materials Chemicals

2010 Rita Moon Financial Bank

My problem is mainly related to the fact that I am dealing with strings. If someone could help me I would appreciate. Thanks a lot for your availability.

Answer 1

[Cedric]

Here one way to proceed:

 [~,ia] = unique( arrayfun( @(r)[C{r,2:end}], 1:size(C,1), 'Unif', 0 )) ;
 C_chg  = C(sort(ia),:)

which outputs

 C_chg = 
    [1997]    'Charles'    'House'     'Materials'    'Chemicals'
    [1997]    'Rita'       'Office'    'Financial'    'Bank'     
    [2000]    'Charles'    'Office'    'Materials'    'Chemicals'
    [2003]    'Rita'       'Star'      'Financial'    'Bank'     
    [2005]    'Charles'    'Castle'    'Materials'    'Chemicals'
    [2010]    'Rita'       'Moon'      'Financial'    'Bank'

when applied to your sample.

[Cedric]

Note that this solution assumes that there is no case where an entry can be obtained by concatenation of several others in other rows. To illustrate, the two following entries are equivalent for my solution above.

    [1997]    'Charles'    'A'     'AAA'    'AA'
    [2000]    'Charles'    'AAA'   'AA'     'A'

I don't beleive that this case can happen, but if you want to avoid this issue, use

[~,ia] = unique( arrayfun( @(r)sprintf('%s ',C{r,2:end}), ...
           1:size(C,1), 'Unif', 0 )) ;

instead of the first line.

[Maria]

Thank you very much. Your code works perfectly!