计算文本列中列表中名称的出现次数

Question

I have a df.Description Column, which has text such as

"shirt XML tag t-shirt (Black) XL",

where the color is always in brackets.

I then have a list of colors:

Colours = ('Blue',
 'Orange',
 'Green',
 'Red',
 'Purple',
 'Brown',
 'Pink',
 'Gray',
 'Olive',
 'Cyan',
 'Black',
 'White') 

My aim is to count the occurrences of each word in the list Colours alongside the word "t-shirt" in that df.Description column.

Basically I need to know which t-shirt colour has been sold the most so I need to count for the colors when there is also the word t-shirt in df.Description .

Answer 1

You can use apply() and value_counts() to calculate the number of occurrance of colors. I would also suggest you to do text preprocessing like converting to small letters and removing punctuations.:

def find_col(x):
  Colours = ['Blue',
 'Orange',
 'Green',
 'Red',
 'Purple',
 'Brown',
 'Pink',
 'Gray',
 'Olive',
 'Cyan',
 'Black',
 'White']

  for col in Colours:
    if (col in x) and ('t-shirt' in x):
      return col + ' t-shirt'
  return 'Not a tshirt'


df['Colors'] = df['Description'].apply(find_col)

df['Colors'].value_counts()