Intelligent E-commerce Recommendation

private double calculatePearsonCorrelation(
        Map<Integer, Double> ratingsA,
        Map<Integer, Double> ratingsB) {

    Set<Integer> commonItems = new HashSet<>(ratingsA.keySet());
    commonItems.retainAll(ratingsB.keySet());

    if (commonItems.size() < 2) return 0.0;

    double sumA = 0, sumB = 0;
    double sumASq = 0, sumBSq = 0, sumAB = 0;
    int n = commonItems.size();

    for (Integer itemId : commonItems) {
        double a = ratingsA.get(itemId);
        double b = ratingsB.get(itemId);
        sumA += a;
        sumB += b;
        sumASq += a * a;
        sumBSq += b * b;
        sumAB += a * b;
    }

    double numerator = sumAB - (sumA * sumB) / n;
    double denominator = Math.sqrt(
        (sumASq - sumA * sumA / n) *
        (sumBSq - sumB * sumB / n)
    );

    return denominator == 0 ? 0.0 : numerator / denominator;
}

for (var simEntry : userSimilarities.entrySet()) {
    double similarity = simEntry.getValue();
    var similarRatings =
        userProductMatrix.get(simEntry.getKey());

    for (var ratingEntry : similarRatings.entrySet()) {
        int productId = ratingEntry.getKey();
        double rating = ratingEntry.getValue();

        if (!targetUserRatings.containsKey(productId)) {
            scores.merge(
                productId, similarity * rating, Double::sum
            );
            simSums.merge(
                productId, Math.abs(similarity), Double::sum
            );
        }
    }
}

return scores.entrySet().stream()
    .filter(e -> simSums.get(e.getKey()) > 0)
    .collect(Collectors.toMap(
        Map.Entry::getKey,
        e -> e.getValue() / simSums.get(e.getKey())
    ))
    .entrySet().stream()
    .sorted(Map.Entry.<Integer, Double>comparingByValue().reversed())
    .limit(maxRecommendations)
    .collect(Collectors.toMap(
        Map.Entry::getKey,
        Map.Entry::getValue,
        (a, b) -> a,
        LinkedHashMap::new
    ));