MediumNLPnlp-041

Build an Inverted Index

Problem

Build an inverted index over docs, a list of token lists.

Return a dictionary mapping each distinct token to the sorted list of document indices that contain it. An empty corpus yields an empty dictionary.

Hint

enumerate gives the document id, and a set per document removes duplicate postings.

case 1

in: [[["a","b"],["b","c"],["a","c"]]]

expected: {"b":[0,1],"a":[0,2],"c":[1,2]}

case 2

in: [[[],["x"]]]

expected: {"x":[1]}

case 3

in: [[]]

expected: {}

case 4

in: [[["the","cat"],["the","dog","cat"]]]

expected: {"cat":[0,1],"the":[0,1],"dog":[1]}

Starter codePython

def inverted_index(docs):
    # Your code here
    pass

4 test cases · run your code against them in the editor