MyMediaLite
3.11
|
k-nearest neighbor user-based collaborative filtering More...
Public Member Functions | |
override void | AddFeedback (ICollection< Tuple< int, int >> feedback) |
Add positive feedback events and perform incremental training More... | |
virtual bool | CanPredict (int user_id, int item_id) |
Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination More... | |
Object | Clone () |
create a shallow copy of the object More... | |
IList< int > | GetMostSimilarUsers (int user_id, uint n=10) |
get the most similar users More... | |
float | GetUserSimilarity (int user_id1, int user_id2) |
get the similarity between two users More... | |
override void | LoadModel (string filename) |
Get the model parameters from a file More... | |
override float | Predict (int user_id, int item_id) |
Predict rating or score for a given user-item combination More... | |
IList< Tuple< int, float > > | Recommend (int user_id, int n=-1, ICollection< int > ignore_items=null, ICollection< int > candidate_items=null) |
Recommend items for a given user More... | |
virtual System.Collections.Generic.IList< Tuple< int, float > > | Recommend (int user_id, int n=-1, System.Collections.Generic.ICollection< int > ignore_items=null, System.Collections.Generic.ICollection< int > candidate_items=null) |
override void | RemoveFeedback (ICollection< Tuple< int, int >> feedback) |
Remove all feedback events by the given user-item combinations More... | |
virtual void | RemoveItem (int item_id) |
Remove all feedback by one item More... | |
virtual void | RemoveUser (int user_id) |
Remove all feedback by one user More... | |
override void | SaveModel (string filename) |
Save the model parameters to a file More... | |
IList< Tuple< int, float > > | ScoreItems (IList< int > accessed_items, IList< int > candidate_items) |
Score a list of items given a list of items that represent a new user More... | |
override string | ToString () |
Return a string representation of the recommender More... | |
override void | Train () |
Learn the model parameters of the recommender from the training data More... | |
Protected Member Functions | |
virtual void | AddItem (int item_id) |
override void | AddUser (int user_id) |
virtual IList< float > | FoldIn (IList< int > items) |
Fold in one user, identified by their items More... | |
void | RecomputeNeighbors (ICollection< int > update_entities) |
void | ResizeNearestNeighbors (int new_size) |
Resizes the nearest neighbors list if necessary More... | |
void | Update (ICollection< Tuple< int, int >> feedback) |
Update the correlation matrix for the given feedback More... | |
Protected Attributes | |
IBinaryDataCorrelationMatrix | correlation_matrix |
Correlation matrix over some kind of entity, e.g. users or items More... | |
uint | k = 80 |
The number of neighbors to take into account for prediction More... | |
IList< IList< int > > | nearest_neighbors |
Precomputed nearest neighbors More... | |
Properties | |
float | Alpha [get, set] |
Alpha parameter for BidirectionalConditionalProbability More... | |
BinaryCorrelationType | Correlation [get, set] |
The kind of correlation to use More... | |
override IBooleanMatrix | DataMatrix [get] |
virtual IPosOnlyFeedback | Feedback [get, set] |
the feedback data to be used for training More... | |
uint | K [get, set] |
The number of neighbors to take into account for prediction More... | |
int | MaxItemID [get, set] |
Maximum item ID More... | |
int | MaxUserID [get, set] |
Maximum user ID More... | |
float | Q [get, set] |
Exponent to be used for transforming the neighbor's weights More... | |
bool | UpdateItems [get, set] |
bool | UpdateUsers [get, set] |
bool | Weighted [get, set] |
Gets or sets a value indicating whether this MyMediaLite.ItemRecommendation.KNN is weighted. More... | |
k-nearest neighbor user-based collaborative filtering
This recommender supports incremental updates for the BinaryCosine and Cooccurrence similarities.
|
inlinevirtual |
Add positive feedback events and perform incremental training
feedback | collection of user id - item id tuples |
Reimplemented from IncrementalItemRecommender.
|
inlinevirtualinherited |
Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination
It is up to the recommender implementor to decide when a prediction is useful, and to document it accordingly.
user_id | the user ID |
item_id | the item ID |
Implements IRecommender.
Reimplemented in ExternalItemRecommender, ExternalRatingPredictor, BiPolarSlopeOne, SlopeOne, Constant, GlobalAverage, UserAverage, ItemAverage, and Random.
|
inlineinherited |
create a shallow copy of the object
|
inlineprotectedvirtual |
Fold in one user, identified by their items
items | the items representing the user |
|
inline |
get the most similar users
user_id | the ID of the user |
n | the number of similar users to return |
Implements IUserSimilarityProvider.
|
inline |
get the similarity between two users
user_id1 | the ID of the first user |
user_id2 | the ID of the second user |
Implements IUserSimilarityProvider.
|
inlinevirtualinherited |
Get the model parameters from a file
filename | the name of the file to read from |
Reimplemented from Recommender.
|
inline |
Predict rating or score for a given user-item combination
user_id | the user ID |
item_id | the item ID |
Implements IRecommender.
|
inherited |
Recommend items for a given user
user_id | the user ID |
n | the number of items to recommend, -1 for as many as possible |
ignore_items | collection if items that should not be returned; if null, use empty collection |
candidate_items | the candidate items to choose from; if null, use all items |
Implemented in WeightedEnsemble, and Ensemble.
|
inlinevirtual |
Remove all feedback events by the given user-item combinations
feedback | collection of user id - item id tuples |
Reimplemented from IncrementalItemRecommender.
|
inlinevirtualinherited |
Remove all feedback by one item
item_id | the item ID |
Implements IIncrementalRecommender.
Reimplemented in BPRMF, BPRSLIM, LeastSquareSLIM, MF, and MostPopular.
|
inlinevirtualinherited |
Remove all feedback by one user
user_id | the user ID |
Implements IIncrementalRecommender.
Reimplemented in LeastSquareSLIM, MF, and MostPopular.
|
inlineprotectedinherited |
Resizes the nearest neighbors list if necessary
new_size | the new size |
|
inlinevirtualinherited |
Save the model parameters to a file
filename | the name of the file to write to |
Reimplemented from Recommender.
|
inline |
Score a list of items given a list of items that represent a new user
accessed_items | the ratings (item IDs and rating values) representing the new user |
candidate_items | the items to be rated |
Implements IFoldInItemRecommender.
|
inlineinherited |
Return a string representation of the recommender
The ToString() method of recommenders should list the class name and all hyperparameters, separated by space characters.
Implements IRecommender.
|
inlinevirtual |
Learn the model parameters of the recommender from the training data
Reimplemented from KNN.
|
inlineprotected |
Update the correlation matrix for the given feedback
feedback | the feedback (user-item tuples) |
|
protectedinherited |
Correlation matrix over some kind of entity, e.g. users or items
|
protectedinherited |
The number of neighbors to take into account for prediction
|
protectedinherited |
Precomputed nearest neighbors
|
getsetinherited |
Alpha parameter for BidirectionalConditionalProbability
|
getsetinherited |
The kind of correlation to use
|
getsetinherited |
the feedback data to be used for training
|
getsetinherited |
The number of neighbors to take into account for prediction
|
getsetinherited |
Maximum item ID
|
getsetinherited |
Maximum user ID
|
getsetinherited |
Exponent to be used for transforming the neighbor's weights
A value of 0 leads to counting of the relevant neighbors. 1 is the usual weighted prediction. Values greater than 1 give higher weight to higher correlated neighbors.
TODO LIT
|
getsetinherited |
Gets or sets a value indicating whether this MyMediaLite.ItemRecommendation.KNN is weighted.
TODO add literature reference