A News Recommendation System Based on Pearson Correlation Coefficient Calculation

The study constructs an architecture using Java EE which improves news distribution systems via Pearson correlation coefficient calculations. Traditional systems have different implementation methods through the combination of behavioral analytics and content relevance modeling beyond using past user behavior data alone. The system constructs two main operational elements that combine news semantic feature analysis with dynamic user preferences using dual-dimensional evaluation methods along with time-expired enhanced Pearson correlation recommendations. güvenlik features enable secure access authentication as well as adaptive suggestion generation while smart news acquisition is integrated into the system. Conclusive experimental analysis based on real-world data demonstrated that the optimized algorithm creates recommendations with 37.2% better accuracy when compared to standard pure collaborative filtering solutions.