The table shows the prediction techniques that are part of the Duine Framework. It also shows which profile models they need, and what adapters they use to obtain input data from domain specific items. The techniques that require an adapter are content dependent prediction techniques. In the current version of the Duine Framework prediction strategies do not need access to any profile model or adapter.
| Prediction Technique | Profile Model | Ratable Item Adapter | Description |
| UserAverage | Rating Model | Returns the average of the ratings that the user has already given to other items | |
| AlreadyKnown | Rating Model | Returns the existing rating of the item for this user (if available) | |
| TopNDeviation | Rating Model | Returns a prediction based on all n ratings from other users that already rated the item | |
| Collaborative Filtering | Rating Model, User Similarity Model | People who have rated the same items the same way in the past probably have the same taste. Based on this knowledge one can predict how much a person likes an unseen item when similar users have already rated that item. | |
| Information Filtering | InterestModel | IRatable2TermWeightAdapter | Extracts information from an item (e.g. the description of a TV program), does a textual analysis and matches the result with the user profile to determine how interesting the item is. |
| Case-based Reasoning | Rating Model, Item Similarity Model | IRatable2SimilarityAdapter | Based on the idea that if two items are similar and if a rating is known for one of them, the rating for the other one will probably be the same. |
| InterestLMS | Interest Model | IRatable2InterestAdapter | Predicts how interesting an item is bases on specific aspects of this item (e.g. the genre) |
