Article

Class dependent feature scaling method using naive Bayes classifier for text datamining

Department of Computer Science, Texas Tech University, Lubbock, TX 79409, USA; Department of Industrial and Systems Engineering and RUTCOR (Rutgers Center for Operations Research), Rutgers, the State University of New Jersey, Piscataway, NJ 08854, USA
Pattern Recognition Letters DOI:10.1016/j.patrec.2008.11.013 pp.477-485
Source: DBLP

ABSTRACT The problem of feature selection is to find a subset of features for optimal classification. A critical part of feature selection is to rank features according to their importance for classification. The naive Bayes classifier has been extensively used in text categorization. We have developed a new feature scaling method, called class–dependent–feature–weighting (CDFW) using naive Bayes (NB) classifier. A new feature scaling method, CDFW–NB–RFE, combines CDFW and recursive feature elimination (RFE). Our experimental results showed that CDFW–NB–RFE outperformed other popular feature ranking schemes used on text datasets.

0 0
 · 
0 Bookmarks
 · 
62 Views
  • Source
    Conference Proceeding: Human-humanoid robot interaction system based on spoken dialogue and vision
    Yanhua Mu, YiXin Yin
    [show abstract] [hide abstract]
    ABSTRACT: In this paper, a natural human-humanoid robot interaction system is proposed. In this system, the type of task is determined with naïve Bayes classifier firstly, and then most of the conditions which are prerequisite to the execution of task are calculated with frame-based reasoning method. The initial position of task object must be calculated with vision. First, the hand position is localized with temporal differencing algorithm. Then, the object region can be determined with the pointing direction of hand. Finally, the reference position of task object in world coordinate system is calculated with binocular vision system.
    Computer Science and Information Technology (ICCSIT), 2010 3rd IEEE International Conference on; 08/2010
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.

Keywords

CDFW
 
CDFW–NB–RFE
 
class–dependent–feature–weighting
 
critical part
 
experimental results
 
feature selection
 
features
 
naive Bayes
 
naive Bayes classifier
 
NB
 
new feature scaling method
 
popular feature ranking schemes
 
rank features
 
recursive feature elimination
 
subset
 
text categorization
 
text datasets