Baselines (recommendation)

Global Average

Global Average uses the average rating value of all ratings to make predictions.

\[\hat { r }_{ ui } = \mu\]

Example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
 import Orange
 from orangecontrib.recommendation import GlobalAvgLearner

 # Load data and train the model
 data = Orange.data.Table('movielens100k.tab')
 learner = GlobalAvgLearner()
 recommender = learner(data)

 prediction = recommender(data[:3])
 print(prediction)
 >>>
 [ 3.52986  3.52986  3.52986]
class orangecontrib.recommendation.GlobalAvgLearner(preprocessors=None, verbose=False)[source]

Global Average

This model takes the average rating value of all ratings to make predictions.

Attributes:
verbose: boolean or int, optional
Prints information about the process according to the verbosity level. Values: False (verbose=0), True (verbose=1) and INTEGER
fit_storage(data)[source]

Fit the model according to the given training data.

Args:
data: Orange.data.Table
Returns:
self: object
Returns self.

User Average

User Average uses the average rating value of a user to make predictions.

\[\hat { r }_{ ui } = \mu_{u}\]

Example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
import Orange
from orangecontrib.recommendation import UserAvgLearner

# Load data and train the model
data = Orange.data.Table('movielens100k.tab')
learner = UserAvgLearner()
recommender = learner(data)

# Make predictions
prediction = recommender(data[:3])
print(prediction)
>>>
[ 3.61538462  3.41304348  3.3515625 ]
class orangecontrib.recommendation.UserAvgLearner(preprocessors=None, verbose=False)[source]

User average

This model takes the average rating value of a user to make predictions.

Attributes:
verbose: boolean or int, optional
Prints information about the process according to the verbosity level. Values: False (verbose=0), True (verbose=1) and INTEGER
fit_storage(data)[source]

Fit the model according to the given training data.

Args:
data: Orange.data.Table
Returns:
self: object
Returns self.

Item Average

Item Average uses the average rating value of an item to make predictions.

\[\hat { r }_{ ui } = \mu_{i}\]

Example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
import Orange
from orangecontrib.recommendation import ItemAvgLearner

# Load data and train the model
data = Orange.data.Table('movielens100k.tab')
learner = ItemAvgLearner()
recommender = learner(data)

# Make predictions
prediction = recommender(data[:3])
print(prediction)
>>>
[ 3.99145299  4.16161616  2.15384615]
class orangecontrib.recommendation.ItemAvgLearner(preprocessors=None, verbose=False)[source]

Item average

This model takes the average rating value of an item to make predictions.

Attributes:
verbose: boolean or int, optional
Prints information about the process according to the verbosity level. Values: False (verbose=0), True (verbose=1) and INTEGER
fit_storage(data)[source]

Fit the model according to the given training data.

Args:
data: Orange.data.Table
Returns:
self: object
Returns self.

User-Item Baseline

User-Item Baseline takes the bias of users and items plus the global average to make predictions.

\[\hat { r }_{ ui } = \mu + b_{u} + b_{i}\]

Example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
 import Orange
 from orangecontrib.recommendation import UserItemBaselineLearner

 # Load data and train the model
 data = Orange.data.Table('movielens100k.tab')
 learner = UserItemBaselineLearner()
 recommender = learner(data)

 # Make predictions
 prediction = recommender(data[:3])
 print(prediction)
 >>>
 [ 4.07697761  4.04479964  1.97554865]
class orangecontrib.recommendation.UserItemBaselineLearner(preprocessors=None, verbose=False)[source]

User-Item baseline

This model takes the bias of users and items plus the global average to make predictions.

Attributes:
verbose: boolean or int, optional
Prints information about the process according to the verbosity level. Values: False (verbose=0), True (verbose=1) and INTEGER
fit_storage(data)[source]

Fit the model according to the given training data.

Args:
data: Orange.data.Table
Returns:
self: object
Returns self.