| | 1 | |
| | 2 | = Things to do = |
| | 3 | |
| | 4 | * '''Adaptation for music domain''' |
| | 5 | * finish skiptrax ontology (see [http://musicontology.com/ Music Ontology]) - 1-3 weeks |
| | 6 | * handle remixes/live/unplugged (song variants) |
| | 7 | * discern song and file - model file ownership |
| | 8 | * Proper GUI with the following functions: |
| | 9 | * metadata importers/exporters (ID3, LastFM, etc.) - 1-2 weeks |
| | 10 | * (adapter for own audio collection) |
| | 11 | * simplify entering song/band information - 1 week |
| | 12 | * (generate/play playlists, possibly with changing style over time (AutoDJ) - 2-6 weeks) |
| | 13 | * '''Metadata-based functionality''' |
| | 14 | * different sharing rights for contacts (e.g. A only ontologies, B also metadata, C everything) - <1 week |
| | 15 | * expert recommender (ask friends about people having knowledge about X/interests in Y) - 1-3 weeks |
| | 16 | * trust visualization |
| | 17 | * scalable algorithms - ''scalable'' = void algo(Graph oldFacts, Graph additionalFacts, Datastructure internalData, Notifications interestingThings) |
| | 18 | * computation of feature summaries - 1 week |
| | 19 | * feature/item similarity measure - 2 weeks |
| | 20 | * structured search (index) |
| | 21 | * implicit trust (= peer competence) metric - 2-5 weeks |
| | 22 | * explicit representation of trust (see Skippies ontology) - 2-5 weeks |
| | 23 | * user interface changes for that |
| | 24 | * trust metric changes for that |
| | 25 | * by combining computed and explicit trust ratings, identifying shared interests/fields of competence should be possible. |
| | 26 | |
| | 27 | |
| | 28 | |
| | 29 | [RelatedWork] |
| | 30 | |
| | 31 | = Ideas = |
| | 32 | |
| | 33 | Skipforward is intended to provide content-bases recommendations, but there might be some cool variants of combining it with collaborative filtering/recommendation: |
| | 34 | |
| | 35 | |
| | 36 | === Collaborative filtering per feature === |
| | 37 | |
| | 38 | Instead of the boring "people who like this also like that" recommendations, this could be adapted in the following way: |
| | 39 | |
| | 40 | People who like A also like B => People who think A has feature F also think that B has feature F[[BR]] |
| | 41 | People who like the same items like you also like this => People who annotated items the same way as you annotated this that way |
| | 42 | |
| | 43 | The last strategy could be used to rate items that haven't been reviewed by the user, but by his friends. |
