An integrated dynamic connection based social group recommendation framework for Netflix to improve data sparsity using novel hybrid filtering techniques comparing with model based conjugate gradient algorithm

Constructing a Dynamic Connection based Group Recommendation Framework for Netflix to improve Data Sparsity and Accuracy percentage using Novel Hybrid Filtering Techniques. Construction of Recommendation System was done by implementing Hybrid Filtering Techniques comparing with memory based Conjugate Gradient algorithm of Sample size is taken as 10. Pre test is calculated with G power of 80% and Threshold 0.05%, Confidence Interval is 95% Mean and Standard Deviation. In this work comparison of Hybrid Filtering Technique and Model Based Conjugate Gradient Algorithm has classified and prediction values from the Recommendations to generate higher accuracy with Hybrid Filtering Technique 91% compared with Model Based Conjugate Gradient Algorithm 70%. In construction of Group Recommendation for Netflix using Hybrid Filtering Techniques gives better Accuracy over Model Based Conjugate Gradient Algorithm.