Ignore:
Timestamp:
03/02/12 12:39:15 (2 years ago)
Author:
janezd <janez.demsar@…>
Branch:
default
Message:

Reorganized documentation about multilabel learning

File:
1 edited

Legend:

Unmodified
Added
Removed
  • Orange/multilabel/mlknn.py

    r9994 r10417  
    66*************************************** 
    77 
    8 ML-kNN Classification is a kind of adaptation method for multi-label classification. 
    9 It is an adaptation of the kNN lazy learning algorithm for multi-label data. 
    10 In essence, ML-kNN uses the kNN algorithm independently for each label :math:`l`. 
    11 It finds the k nearest examples to the test instance and considers those that are 
    12 labeled at least with :math:`l` as positive and the rest as negative. 
    13 Actually this method follows the paradigm of Binary Relevance (BR). What mainly 
    14 differentiates this method from BR is the use of prior probabilities. ML-kNN has also 
    15 the capability of producing a ranking of the labels as an output. 
    16 For more information, see Zhang, M. and Zhou, Z. 2007. `ML-KNN: A lazy learning 
    17 approach to multi-label learning <http://dx.doi.org/10.1016/j.patcog.2006.12.019>`_.  
    18 Pattern Recogn. 40, 7 (Jul. 2007), 2038-2048.   
     8ML-kNN Classification is an adaptation kNN for multi-label 
     9classification.  In essence, ML-kNN uses the kNN algorithm 
     10independently for each label :math:`l`.  It finds the k nearest 
     11examples to the test instance and considers those that are labeled at 
     12least with :math:`l` as positive and the rest as negative.  What 
     13mainly differentiates this method from other binary relevance (BR) 
     14methods is the use of prior probabilities. ML-kNN can also rank labels. 
     15 
     16For more information, see Zhang, M. and Zhou, Z. 2007. `ML-KNN: A lazy 
     17learning approach to multi-label learning 
     18<http://dx.doi.org/10.1016/j.patcog.2006.12.019>`_.  Pattern 
     19Recogn. 40, 7 (Jul. 2007), 2038-2048. 
    1920 
    2021.. index:: ML-kNN Learner 
     
    5051class MLkNNLearner(_multiknn.MultikNNLearner): 
    5152    """ 
    52     Class implementing the ML-kNN (Multi-Label k Nearest Neighbours) algorithm. The class is based on the  
    53     pseudo-code made available by the authors. 
     53    Class implementing the ML-kNN (Multi-Label k Nearest Neighbours) 
     54    algorithm. The class is based on the pseudo-code made available by 
     55    the authors. 
    5456     
    5557    The pseudo code of ML-kNN: 
Note: See TracChangeset for help on using the changeset viewer.