Feature Selection for Social Media Data

2014 ◽  
Vol 8 (4) ◽  
pp. 1-27 ◽  
Author(s):  
Jiliang Tang ◽  
Huan Liu
Keyword(s):  
Social Media ◽  
Feature Selection ◽  
Social Media Data ◽  
Selection For ◽  
Media Data

Optimizing Social Media Data Using Genetic Algorithm

2019 ◽  
pp. 126-153
Author(s):  
Sourav Das ◽  
Anup Kumar Kolya ◽  
Dipankar Das
Keyword(s):  
Genetic Algorithm ◽  
Social Media ◽  
Feature Selection ◽  
Feature Reduction ◽  
Social Media Data ◽  
Training Time ◽  
Redundant Feature ◽  
Media Data ◽  
And Training

Twitter-based research for sentiment analysis is popular for quite some time now. This is used to represent documents in a corpus usually. This increases the time of classification and also increases space complexity. It is hence very natural to say that non-redundant feature reduction of the input space for a classifier will improve the generalization property of a classifier. In this approach, the researchers have tried to do feature selection using Genetic Algorithm (GA) which will reduce the set of features into a smaller subset. The researchers have also tried to put forward an approach using Genetic Algorithm to reduce the modelling complexity and training time of classification algorithm for 10k Twitter data based on GST. They aim to improve the accuracy of the classification that the researchers have obtained in a preface work to this work and achieved an accuracy of 87% through this work. Hence the Genetic Algorithm will do the feature selection to reduce the complexity of the classifier and give us a better accuracy of the classification of the tweet.

scholarly journals Semi-Supervised Feature Selection with Universum Based on Linked Social Media Data

2014 ◽  
Vol E97.D (9) ◽  
pp. 2522-2525
Author(s):  
Junyang QIU ◽  
Yibing WANG ◽  
Zhisong PAN ◽  
Bo JIA
Keyword(s):  
Social Media ◽  
Feature Selection ◽  
Social Media Data ◽  
Media Data
Sign in / Sign up

Export Citation Format

Share Document