Pearson Correlation in Recommendation Systems

Pearson Correlation in Recommendation Systems | CodeSignal Learn

Introduction to Similarity Measures in Recommendation Systems

In the world of recommendation systems, one of the keys to success is understanding the similarity between users or items. This understanding forms the backbone of making accurate recommendations. Similarity measures allow us to identify users with similar preferences, improving the quality and relevance of recommendations. By exploring these measures, we can enhance the effectiveness of our recommendation algorithms.

Recap of Essential Setup Steps

We will be working with user rating datasets, which we will represent as vectors in this lesson. Here's a simple code block to demonstrate setting up and using std::vector to create user rating datasets: C++#include <vector> // Example user ratings std::vector<double> user1_ratings = {5, 3, 4, 2, 1, 5, 3, 4, 2, 1, 5, 3, 4, 2, 1}; std::vector<double> user2_ratings = {5, 3, 4, 5, 1, 3, 3, 4, 2, 1, 5, 2, 2, 2, 1}; Each index in user1_ratings and user2_ratings corresponds to the rating of the same item by both users. These vectors can be extracted from the user-item matrix, but this time we will simply define them like this for brevity. If a rating is missing for one user in the user-item matrix, that item should be excluded from the calculation. This ensures that only ratings for items both users have rated are compared, which is essential for accurate similarity measurement.

Understanding Pearson Correlation

r = \frac{\sum (x_i - \bar{x}) (y_i - \bar{y})}{\sqrt{\sum (x_i - \bar{x})^2 \sum (y_i - \bar{y})^2}}

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Pearson Correlation measures the strength and direction of a linear relationship between two sets of data. It's a popular tool in recommendation systems because it helps gauge the similarity between users based on their rating trends, rather than their absolute ratings.

The formula for Pearson Correlation is:

$r = \frac{\sum (x_i - \bar{x}) (y_i - \bar{y})}{\sqrt{\sum (x_i - \bar{x})^2 \sum (y_i - \bar{y})^2}}$

$x_i$ , $y_i$ are the individual ratings.
$\bar{x}$ , $\bar{y}$ are the mean ratings for each user.
The numerator sums the product of the differences from the mean.
The denominator normalizes this sum with the square root of the squared differences.

The Pearson correlation coefficient ranges from -1 to 1:

A coefficient of 1 indicates a perfect positive linear relationship.
A coefficient of -1 indicates a perfect negative linear relationship.
A coefficient of 0 indicates no linear correlation.

Understanding this formula is crucial for implementing the measure and interpreting its results in the context of user similarity.

Step-by-step Implementation

Let’s break down the implementation of the Pearson Correlation function to understand how to calculate it step by step. C++ #include <vector> #include <numeric> #include <cmath> // Function to calculate Pearson correlation between two users double pearsonCorrelation(const std::vector<double>& x, const std::vector<double>& y) {int n = x.size(); double meanX = std::accumulate(x.begin(), x.end(), 0.0) / n; double meanY = std::accumulate(y.begin(), y.end(), 0.0) / n; double numerator = 0.0, denomX = 0.0, denomY = 0.0; for (int i = 0; i < n; ++i) {double diffX = x[i] - meanX; double diffY = y[i] - meanY; numerator += diffX * diffY; denomX += diffX * diffX; denomY += diffY * diffY;} double denominator = std::sqrt(denomX * denomY); return (denominator == 0.0) ? 0.0 : numerator / denominator;} #include <vector> #include <numeric> #include <cmath> // Function to calculate Pearson correlation between two users double pearsonCorrelation(const std::vector<double>& x, const std::vector<double>& y) {int n = x.size(); double meanX = std::accumulate(x.begin(), x.end(), 0.0) / n; double meanY = std::accumulate(y.begin(), y.end(), 0.0) / n; double numerator = 0.0, denomX = 0.0, denomY = 0.0; for (int i = 0; i < n; ++i) {double diffX = x[i] - meanX; double diffY = y[i] - meanY; numerator += diffX * diffY; denomX += diffX * diffX; denomY += diffY * diffY;} double denominator = std::sqrt(denomX * denomY); return (denominator == 0.0) ? 0.0 : numerator / denominator;} We start by calculating the mean rating for each user, then determine how each rating deviates from the mean, and finally compute the numerator and denominator for the correlation formula. This process ensures that the similarity measure reflects the trends in user ratings rather than their absolute values.

Example Application and Interpretation

Let's apply this function to our previous example user ratings and see how it works in practice. C++// (Assume pearsonCorrelation function is defined as above) int main() { // Example user ratings for commonly rated items std::vector<double> user1_ratings = {5, 3, 4, 2, 1, 5, 3, 4, 2, 1, 5, 3, 4, 2, 1}; std::vector<double> user2_ratings = {5, 3, 4, 5, 1, 3, 3, 4, 2, 1, 5, 2, 2, 2, 1}; double pearson_similarity = pearsonCorrelation(user1_ratings, user2_ratings); std::cout << "Pearson Correlation: " << pearson_similarity << std::endl; return 0; }// (Assume pearsonCorrelation function is defined as above) int main() { // Example user ratings for commonly rated items std::vector<double> user1_ratings = {5, 3, 4, 2, 1, 5, 3, 4, 2, 1, 5, 3, 4, 2, 1}; std::vector<double> user2_ratings = {5, 3, 4, 5, 1, 3, 3, 4, 2, 1, 5, 2, 2, 2, 1}; double pearson_similarity = pearsonCorrelation(user1_ratings, user2_ratings); std::cout << "Pearson Correlation: " << pearson_similarity << std::endl; return 0; } Output: Pearson Correlation: 0.7Pearson Correlation: 0.7 In this context, a Pearson Correlation of 0.7 indicates a positive relationship between the rating trends of the two users. This suggests that they have similar interests, making it easier to recommend new items they might both enjoy.

Understanding Correlation with Additional Example

The term correlation refers to the statistical measure that describes the extent to which two variables change together. For instance, even if two users rate more items in common, their correlation might not be as high as that of another pair because correlation focuses on how the ratings move together. Let's illustrate with a third user: C++// (Assume pearsonCorrelation function is defined as above) int main() { // Additional user ratings std::vector<double> user1_ratings = {2, 2, 3, 4, 4}; std::vector<double> user2_ratings = {2, 5, 3, 1, 4}; std::vector<double> user3_ratings = {1, 1, 2, 4, 3}; // Calculate Pearson correlations double pearson_similarity_12 = pearsonCorrelation(user1_ratings, user2_ratings); double pearson_similarity_13 = pearsonCorrelation(user1_ratings, user3_ratings); std::cout << "Pearson Correlation between User 1 and User 2: " << pearson_similarity_12 << std::endl; // -0.32 std::cout << "Pearson Correlation between User 1 and User 3: " << pearson_similarity_13 << std::endl; // 0.96 return 0; }// (Assume pearsonCorrelation function is defined as above) int main() { // Additional user ratings std::vector<double> user1_ratings = {2, 2, 3, 4, 4}; std::vector<double> user2_ratings = {2, 5, 3, 1, 4}; std::vector<double> user3_ratings = {1, 1, 2, 4, 3}; // Calculate Pearson correlations double pearson_similarity_12 = pearsonCorrelation(user1_ratings, user2_ratings); double pearson_similarity_13 = pearsonCorrelation(user1_ratings, user3_ratings); std::cout << "Pearson Correlation between User 1 and User 2: " << pearson_similarity_12 << std::endl; // -0.32 std::cout << "Pearson Correlation between User 1 and User 3: " << pearson_similarity_13 << std::endl; // 0.96 return 0; } Despite having more common ratings with User 2, User 1’s ratings' movement is more aligned with User 3’s. This illustrates how correlation is about the trend in ratings rather than the sheer count of mutual ratings. This fact is not an advantage or disadvantage of this similarity measure, but it is a fact that you should be aware of when using it. There are other metrics that do not have this feature, like cosine similarity.