A basic recommendation system based on loans data

lis_advent_calendar-recsys.py

from math import sqrt
import json
import pathlib
import pprint


def fill_resources(patrons):
    all_resources = {}
    for patron in patrons:
        for resource, rating in patrons[patron].items():
            all_resources[resource] = 1

    for ratings in patrons.values():
        for item in all_resources:
            if item not in ratings:
                ratings[item] = 0.0


def sim_pearson(loans, p1, p2):
    si = {}
    for item in loans[p1]:
        if item in loans[p2]:
            si[item] = 1

    n = len(si)

    if n == 0:
        return 0

    sum1 = sum([loans[p1][it] for it in si])
    sum2 = sum([loans[p2][it] for it in si])

    sum1Sq = sum([pow(loans[p1][it], 2) for it in si])
    sum2Sq = sum([pow(loans[p2][it], 2) for it in si])

    pSum = sum([loans[p1][it] * loans[p2][it] for it in si])

    num = pSum - (sum1 * sum2 / n)
    den = sqrt((sum1Sq - pow(sum1, 2) / n) * (sum2Sq - pow(sum2, 2) / n))

    if den == 0:
        return 0
    r = num / den

    return r


def getRecommendations(loans, patron, similarity=sim_pearson):
    totals = {}
    simSums = {}

    for other in loans:
        if other == patron:
            continue

        sim = similarity(loans, patron, other)

        if sim <= 0:
            continue

        for item in loans[other]:
            if item not in loans[patron] or loans[patron][item] == 0:
                totals.setdefault(item, 0)
                totals[item] += loans[other][item] * sim

                simSums.setdefault(item, 0)
                simSums[item] += sim

    rankings = [(total / simSums[item], item) for item, total in totals.items()]
    rankings.sort()
    rankings.reverse()

    return rankings


if __name__ == "__main__":
    patrons = json.loads(pathlib.Path("dataset.json").read_text())
    fill_resources(patrons)
    user = "DI PRETORO, Emmanuel"
    pprint.pprint(getRecommendations(patrons, user)[0:10])