2015 Science and Information Conference (SAI) 2015
DOI: 10.1109/sai.2015.7237148
|View full text |Cite
|
Sign up to set email alerts
|

Metaheuristic algorithms for feature selection in sentiment analysis

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
18
0
1

Year Published

2017
2017
2020
2020

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 55 publications
(19 citation statements)
references
References 33 publications
0
18
0
1
Order By: Relevance
“…The increasing of b value improves recognition effect to some extent. In view of the recognition conditions for two corpuses, b value is with better effect while increasing in the interval of [5,30], but the improvement of performance is not significant if continue to increase b value, which explains that b=30 is a turning point for identifying the stability of performance. Moreover, generation number L of subspace has greater influence on recognizing the performance of model.…”
Section: Experimental Results and Analysismentioning
confidence: 99%
See 2 more Smart Citations
“…The increasing of b value improves recognition effect to some extent. In view of the recognition conditions for two corpuses, b value is with better effect while increasing in the interval of [5,30], but the improvement of performance is not significant if continue to increase b value, which explains that b=30 is a turning point for identifying the stability of performance. Moreover, generation number L of subspace has greater influence on recognizing the performance of model.…”
Section: Experimental Results and Analysismentioning
confidence: 99%
“…(2) For the selection of sentiment characteristics, firstly, it adopts word segmentation tool ICTCLAS2012 [5] of Chinese Academy of Sciences to make batch word segmentation for text set, saves word segmentation results and then extracts sentiment words included in word segmentation results as characteristics based on sentiment dictionary made based on scheme (1). ( based on linear kernel function for its extensive application in text classification and good classification performance and processing speed in sparse feature space.…”
Section: Problem Descriptionmentioning
confidence: 99%
See 1 more Smart Citation
“…The IMDb is a large database with related and comprehends information on movies-past, present and future [18]. It commenced as shell script set and data file.…”
Section: Internet Movie Database (Imdb) Datasetmentioning
confidence: 99%
“…Thither are numerous approaches (including machine learning approach, lexicon based approach) for calculating sentiment. [4] Prefers the metaheuristic algorithm which is used as feature selection in sentiment analysis. This algorithmis used for selecting the optimum features from the customer reviews.…”
mentioning
confidence: 99%