Natural language processing based features for sarcasm detection: An investigation using bilingual social media texts

Suhaimin, Mohd Suhairi Md; Hijazi, Mohd Hanafi Ahmad; Alfred, Rayner; Coenen, Frans

doi:10.1109/icitech.2017.8079931

Cited by 38 publications

(18 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sentiment analysis could be adversely affected by the occurrence of sarcastic elements (Suhairi et al [15]). Hence, the authors appraised the effectiveness of five categories of NLP-based features in detecting sarcasm in bilingual social media text.…”

Section: Sarcasm Detectionmentioning

confidence: 99%

Challenges and Development in Malay Natural Language Processing

2020

jcr

View full text Add to dashboard Cite

This paper aims to review thechallenges and current development in the Malay Natural Language Processing (NLP). The incongruence of the spoken and written Malay language presents a NLP challenge and is attributable to both (1) the cultural, state level dialects and (2) the mixed usage of English-Malay languages, being part of the South East Asia region's colonization legacy. There is also inadequate publicly available Malay NLP resources in the form of domain specific text corpus, open source Part of Speech (POS) and Named Entity Recognition (NER) systems. That said, significant progress is being made in Malay lexicon, information retrieval (stop words and NER), text normalization, machine translation, sentiment analysis and test summarization. These present potential commercial opportunity in reaching out to the 215 million native speakers in the region, not counting theglobal Malay diaspora community. The Malay NLP is also crucial in protecting national and cross border as digital counter-terrorism arsenal.

show abstract

Section: Sarcasm Detectionmentioning

confidence: 99%

Challenges and Development in Malay Natural Language Processing

2020

jcr

View full text Add to dashboard Cite

show abstract

“…), Tanda seru (! ), Kutipan (") dan "'), tagar (#), dan emotikon [13]. selain kalimat negasi seperti "no", "none", "never" dan lain-lain.…”

Section: Pendahuluanunclassified

Studi Komparatif Metode Ekstraksi Fitur pada Analisis Sentimen Maskapai Penerbangan Menggunakan Support Vector Machine dan Maximum Entropy

Cindo

Rini

Ermatita

2019

RESTI

View full text Add to dashboard Cite

Almost all companies use social media to improve their product services and provide after-sales services that allow their customers to review the quality of their products. By using Twitter social media to be an important source for tracking sentiment analysis. Sentiment analysis is one of the most popular studies today, using sentiment analysis companies can analyze customer satisfaction to improve their services. This study aims to analyze airline sentiments with five different features such as pragmatic, lexical n-gram, POS, sentiment, and LDA using the Support Vector Machine and Maximum Entropy methods. The best results can be obtained using the Maximum Entropy method using all feature extraction with an accuracy of 92.7% and in the Support Vector Machine method, the accuracy obtained is 89.2%.

show abstract

“…Early work on sarcasm detection on Twitter data using punctuation and interjection successfully gained a fmeasure score of 0.813 [3]. In another work which detects sarcasm in Facebook comment posts, combination of interjection and punctuation with syntactic feature increased the f-measure score into 0.852 [4]. Despite many researches have been conducted to detect sarcasm in English, there is only one of a kind on Indonesian.…”

Section: Related Workmentioning

confidence: 99%

Sarcasm Detection on Indonesian Twitter Feeds

Rahayu¹,

Kuntur²,

Hayatin³

2018

EECSI

View full text Add to dashboard Cite

In social media, some people use positive words to express negative opinion on a topic which is known as sarcasm. The existence of sarcasm becomes special because it is hard to be detected using simple sentiment analysis technique. Research on sarcasm detection in Indonesia is still very limited. Therefore, this research proposes a technique in detecting sarcasm in Indonesian Twitter feeds particularly on several critical issues such as politics, public figure and tourism. Our proposed technique uses two feature extraction methods namely interjection and punctuation. These methods are later used in two different weighting and classification algorithms. The empirical results demonstrate that combination of feature extraction methods, tf-idf, k-Nearest Neighbor yields the best performance in detecting sarcasm.

show abstract

Natural language processing based features for sarcasm detection: An investigation using bilingual social media texts

Cited by 38 publications

References 16 publications

Challenges and Development in Malay Natural Language Processing

Challenges and Development in Malay Natural Language Processing

Studi Komparatif Metode Ekstraksi Fitur pada Analisis Sentimen Maskapai Penerbangan Menggunakan Support Vector Machine dan Maximum Entropy

Sarcasm Detection on Indonesian Twitter Feeds

Contact Info

Product

Resources

About