基于字典的关键字分类

Question

I'm pretty new to programming and have been pretty enthralled by its power so far. In this vein, there was a problem I had in which have a dataset in which one of the variable is a commodity name: "apple", "pear", "cauliflower", "clog", "sneaker", etc. I want to try and group the commodities into something a little more high-order: "fruits", "vegetables"," "shoes", etc. My sense from doing some searching is that this would be a dictionary-based chunking problem, but I'm not sure how to implement a solution. I could get lists of vegetables, fruits, and types of shoes pretty easily, but are there existing packages that could help with this kind of a problem specifically? I'm most comfortable with Python and R, so anything that can be used with those languages would be most helpful.

Apologies if this question isn't written in a specific-enough way. I'm new to stackoverflow and am still getting the hang of the thing.

Clarification : I'm trying to create a new dataset with these new higher-order labels.

Answer 1

Here's how I would do it:

higher_order_conversion = {
    ('apple', 'pear', 'kiwi'): 'fruit',   #the keys must be tuples, not lists 
    ('X', 'Y', 'Z'): 'letter', # (because tuples are immutable and therefore hashable)
    ('loafers', 'sneakers', 'high heels'): 'shoes'
}

data_set = [[125, 'apple'], #these numbers are id numbers, or whatever extra information you might have packaged with your data
            [126, 'Y'],
            [127, 'loafers'],
            [103, 'kiwi']
            ]

print 'before', data_set

for data in data_set:
    for lower_order_list in higher_order_conversion.keys():
        if data[1] in lower_order_list:
            data[1] = higher_order_conversion[lower_order_list]

print 'after', data_set

Output:

before [[125, 'apple'], [126, 'Y'], [127, 'loafers'], [103, 'kiwi']]
after [[125, 'fruit'], [126, 'letter'], [127, 'shoes'], [103, 'fruit']]

Hopefully this gives you some ideas.