MATLAB: Find Data in Table

columnsdatafindMATLABqualitativerowssorttable

As a follow up question, my data is also inconsistent in how it is described. For example items 10 and 11 in my new table:
11×6 table
    ID    Speed    Time    Fruit1      Fruit2        Fruit3   
    __    _____    ____    _______    ________    ____________
     1    3        9       'Apple'    'Jam'       'Cake'      
     2    6        5       'Cake'     'Orange'    'Jam'       
     3    3        4       'Jam'      'Cake'      'Apple'     
     4    5        2       'Jam'      'Orange'    'Cake'      
     5    9        4       'Jam'      'Jam'       'Orange'    
     6    5        3       'Pear'     'Jam'       'Cake'      
     7    3        5       'Cake'     'Cake'      'Cake'      
     8    2        3       'Jam'      'Cake'      'Apple'     
     9    5        3       'Jam'      'Orange'    'Jam'       
    10    7        1       'Jam'      'Jam'       'JamApple'  
    11    1        4       'Jam'      'Jam'       'OrangeCake'
Is it possible to include the JamApple as a type of Apple please?

Best Answer

Using categorical for the string columns has some advantages. I fixed up your listing above to be able to read as a table via
> t=readtable('cox.dat');
>> whos t
  Name       Size            Bytes  Class    Attributes
  t         10x6              2848  table              
>> t.Properties
ans = 
             Description: ''
    VariableDescriptions: {}
           VariableUnits: {}
          DimensionNames: {'Row'  'Variable'}
                UserData: []
                RowNames: {}
           VariableNames: {'ID'  'Speed'  'Time'  'Fruit1'  'Fruit2'  'Fruit3'}
>>
With that it's easy enough to write
   >> ix=any(ismember(t{:,4:6},{'Apple';'Orange'}),2);  % find any row with apple or orange
  >> t(ix,:)                                            % display those found
  ans = 
      ID    Speed    Time    Fruit1    Fruit2    Fruit3
      __    _____    ____    ______    ______    ______
       1    3        9       Apple     Jam       Cake  
       2    6        5       Cake      Orange    Jam   
       3    3        4       Jam       Cake      Apple 
       4    5        2       Jam       Orange    Cake  
       5    9        4       Jam       Jam       Orange
       8    2        3       Jam       Cake      Apple 
       9    5        3       Jam       Orange    Jam   
      10    7        9       Jam       Jam       Apple 
  >>
Use the logical index vector to access any other variable of interest.
It would be better going forward, of course, to normalize the database to put the three disparate properties in their own variable.
Related Question