2017
DOI: 10.1007/978-3-319-71273-4_22
|View full text |Cite
|
Sign up to set email alerts
|

Stance Classification of Tweets Using Skip Char Ngrams

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
21
0
1

Year Published

2018
2018
2023
2023

Publication Types

Select...
3
3
2

Relationship

2
6

Authors

Journals

citations
Cited by 21 publications
(22 citation statements)
references
References 15 publications
0
21
0
1
Order By: Relevance
“…We have submitted 17 models: 6 models to task 6-A, 6 models to task 6-B, and 5 models to task 6-C. We applied the Python module called Scikitlearn (Pedregosa et al, 2011) using the TF-IDF scheme called TfidfTransformer 3 and we applied various supervised ML methods with various numbers of n-gram features, skip word/char ngrams (HaCohen-Kerner et al, 2017) and combinations of pre-processing types.…”
Section: The Submitted Models and Experimental Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…We have submitted 17 models: 6 models to task 6-A, 6 models to task 6-B, and 5 models to task 6-C. We applied the Python module called Scikitlearn (Pedregosa et al, 2011) using the TF-IDF scheme called TfidfTransformer 3 and we applied various supervised ML methods with various numbers of n-gram features, skip word/char ngrams (HaCohen-Kerner et al, 2017) and combinations of pre-processing types.…”
Section: The Submitted Models and Experimental Resultsmentioning
confidence: 99%
“…Stance classification of tweets was investigated by HaCohen-Kerner et al (2017). Given test datasets of tweets from five various topics, they classified the stance of the tweet authors as either in FAVOR of the target, AGAINST it, or NONE.…”
Section: Tweet Classificationmentioning
confidence: 99%
“…Bu sebeple baz yöntem olarak kullanılmaktadır. Sıklıkla kullanılan diğer yöntemler arasında NB (HaCohen-Kerner et al, 2017;Lai et al, 2016), ESA (Hercig et al, 2017;Zhang et al, 2017;Zhou et al, 2017), RO Algoritması (Aker et al, 2017;Tsakalidis et al, 2018) yer almaktadır.…”
Section: İlgili çAlışmalarunclassified
“…In this experiment we tested the addition of open n-grams to subword features, to see if they improved fastText vector representations. We experimented many different encoding schemes, and found that including both open, and regular contiguous n-grams gave inferior results to just using open n-grams 8 . In other words, contiguous n-grams always degraded performance.…”
Section: Experiments 2: Non-contiguous N-gramsmentioning
confidence: 99%