Item-Based Cosine Predict

Problem

Predict a rating as the similarity-weighted average of the user's ratings on items similar to the target: sum(sim * rating) / sum(|sim|) over items present in both dicts.

Return 0.0 when there is no overlap or the weight sum is 0.

Hint

Item-item similarity weights the user's own ratings of related items.

Test cases

case 1

in: [{"a":5,"b":3},{"a":0.8,"b":0.4}]

expected: 4.333333333333333

case 2

in: [{"a":5,"b":1},{"a":0.5,"b":-0.5}]

expected: 2

case 3

in: [{"a":5},{"b":0.5}]

expected: 0

case 4

in: [{},{"a":0.5}]

expected: 0

Starter codePython

def item_based_cosine_predict(user_ratings, similarities):
    # Your code here
    pass

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