Find frequency of words from different books

I have a cell array of data collected from 5 different books

This is one of the cell arrays. It gives me the count of each word in the book (I used count{ii} = tabulate(text{ii}) ).

I need to create a unique count for all the words found in all the 5 books. So, for example, for the word 'the', I have to sum up all the frequencies in all 5 cells.

I was thinking about using a table but I really can't get it done.

Any ideas?

Answer 1

[Voss]

Maybe this will help:

% example data:
counts = { ...
    {'the' 464; 'project' 87; 'of' 253} ...
    {'the' 300; 'of' 314; 'nothing' 17; 'project' 13} ...
    {'the' 100; 'price' 99; 'of' 114; 'everything' 12; 'value' 88; 'nothing' 54} ...
    }

counts = 1x3 cell array
    {3x2 cell}    {4x2 cell}    {6x2 cell}

% concatenate the cell arrays in counts and convert into a table
T = cell2table(vertcat(counts{:}),'VariableNames',{'word','count'})

T = 13x2 table
         word         count
    ______________    _____

    {'the'       }     464 
    {'project'   }      87 
    {'of'        }     253 
    {'the'       }     300 
    {'of'        }     314 
    {'nothing'   }      17 
    {'project'   }      13 
    {'the'       }     100 
    {'price'     }      99 
    {'of'        }     114 
    {'everything'}      12 
    {'value'     }      88 
    {'nothing'   }      54 

% use groupsummary to find the total counts
G = groupsummary(T,'word','sum')

G = 7x3 table
         word         GroupCount    sum_count
    ______________    __________    _________

    {'everything'}        1             12   
    {'nothing'   }        2             71   
    {'of'        }        3            681   
    {'price'     }        1             99   
    {'project'   }        2            100   
    {'the'       }        3            864   
    {'value'     }        1             88   

[L]

That is exacly what I needed. What does the column Group Count means?

[Voss]

You're welcome!

GroupCount is the number of times each word appears in the table T, so that would correspond to the number of books each word appears in. I don't think you need that information (it's automatically included by groupsummary), and you can remove it.

% example data:
counts = { ...
    {'the' 464; 'project' 87; 'of' 253} ...
    {'the' 300; 'of' 314; 'nothing' 17; 'project' 13} ...
    {'the' 100; 'price' 99; 'of' 114; 'everything' 12; 'value' 88; 'nothing' 54} ...
    };

% concatenate the cell arrays in counts and convert into a table
T = cell2table(vertcat(counts{:}),'VariableNames',{'word','count'});

% use groupsummary to find the total counts
G = groupsummary(T,'word','sum');

% remove GroupCount
G = removevars(G,'GroupCount')

G = 7x2 table
         word         sum_count
    ______________    _________

    {'everything'}        12   
    {'nothing'   }        71   
    {'of'        }       681   
    {'price'     }        99   
    {'project'   }       100   
    {'the'       }       864   
    {'value'     }        88