|
| S. no. | Paper reference | Techniques used | Key findings | Observations |
|
| 1 | [10] | Probabilistic factor analysis | Determines authenticity and the realistic nature of news | Calculates the probability of news from news data sets collected by different websites |
| 2 | [11] | Collaborative filtering | Describes profile similarity and trust-based recommendation systems | A correlation between trust and the largest single difference in score exists along with overall similarity |
| 3 | [12] | Collaborative filtering | Describes social networks based on semantic networks and trust for movie recommendations | Creates predictive rating recommendations for movies |
| 4 | [13] | KNN | Includes multigraph ranking model | Results represent different user relationships in multiple graphs and recommend the nearest neighbors of specific users |
| 5 | [14] | Probabilistic factor analysis | Includes trust-based recommendations in a social network | It has lower recommendation probabilities |
| 6 | [15] | KNN | Includes a hybrid approach | Combines user ratings and social trust. Compared with other trust-aware recommendation work, their method uses untrusted links and investigates their dissemination effect |
| 7 | [16] | Collaborative filtering | Includes user relevance and evolutionary clustering | Correlation is calculated by combining user satisfaction and potential score information |
| 8 | [17] | Collaborative filtering | Includes a new similarity calculation method called JacRA | Calculates selection of the items and the ratings. It has complex calculations to determine ratings |
| 9 | [18] | Collaborative filtering | It solves data scarcities, cold start, recommendation accuracy, and timelines as an improved collaborative recommendation algorithm | It involves the traditional similarity of the collaborative recommendation algorithm, for an improved recommendation |
| 10 | [19] | Matrix factorization | Includes a hybrid approach combining social behavior, movie genre, and existing collaborative filtering algorithms | Calculates similarity of movies to predict user ratings |
| 11 | [20] | Collaborative filtering | Includes collaborative filtering based on the ratings of the movies implemented by Apache mahout | Considers user ratings to recommend movies |
| 12 | [21] | Matrix factorization | Includes model-based approach using matrix factorization techniques in social networks | It resolves the cold start problem to some extent using social MF |
| 13 | [22] | Probabilistic factor analysis | Includes probabilistic factor analysis method that calculates multifaceted trust relationships and user profiles by sharing the user’s potential feature space | It cannot generate better results using trust relations for predictions |
| 14 | [23] | Content-based, collaborative filtering | It includes a collaborative filtering approach and uses the information provided by users | It provides suggestions to the users using the two renowned algorithms |
| 15 | [24] | Matrix factorization | Includes dual role preferences (trustee/trustee specific preferences), and trust-aware recommendations are achieved by modeling explicit interactions | Using explicit interactions makes it difficult to compute due to privacy issues |
|
