Stierlitz Meets SVM: Humor Detection in Russian

Ermilov, Anton; Murashkina, Natasha; Goryacheva, Valeria; Braslavski, Pavel

doi:10.1007/978-3-030-01204-5_17

Cited by 12 publications

(19 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We tested the obtained models on held-out test sets of STIER-LITZ and FUN, as well as on smaller manually annotated GOLD and PUNS collections. In addition, we were able to apply the best model from (Ermilov et al, 2018) to the test data. Table 2 summarizes performance of the models.…”

Section: Resultsmentioning

confidence: 99%

“…3 Data STIERLITZ and PUNS. We started with a dataset of Russian one-liners and non-humorous texts used previously in (Ermilov et al, 2018). The balanced dataset was assembled by complementing a collection of jokes from social media (Bolotova et al, 2017) with non-humorous proverbs, news headlines, and sentences from fiction books.…”

Section: Related Workmentioning

confidence: 99%

“…Following the authors, we refer to the dataset as STIERLITZ. 2 We also use a small collection of Russian puns from (Ermilov et al, 2018) for evaluation. Puns as a special type of humor seem to be less articulated in the Russian culture compared to 2 Stierlitz is a protagonist of a popular TV series, a Soviet spy working undercover in Nazi Germany.…”

Section: Related Workmentioning

confidence: 99%

“…We thank volunteers who took part in data annotation. We are also thanksful to authors of (Ermilov et al, 2018), who ran their code on our data.…”

Section: Acknowledgmentsmentioning

confidence: 99%

“…Humor recognition is usually formulated as a classification problem with a wide variety of features -syntactic parsing, alliteration and rhyme, antonymy and other WordNet relations, dictionaries of slang and sexually explicit words, polarity and subjectivity lexicons, distances between words in terms of word2vec representations, word association measures, etc. (Taylor and Mazlack, 2004;Mihalcea and Strapparava, 2005;Kiddon and Brun, 2011;Yang et al, 2015;Zhang and Liu, 2014;Liu et al, 2018;Cattle and Ma, 2018;Ermilov et al, 2018). A cognate task is humor ranking (Shahaf et al, 2015;Potash et al, 2017).…”

Section: Related Workmentioning

confidence: 99%

See 4 more Smart Citations

Large Dataset and Language Model Fun-Tuning for Humor Recognition

Blinov¹,

Bolotova-Baranova

Braslavski³

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Self Cite

View full text Add to dashboard Cite

The task of humor recognition has attracted a lot of attention recently due to the urge to process large amounts of user-generated texts and rise of conversational agents. We collected a dataset of jokes and funny dialogues in Russian from various online resources and complemented them carefully with unfunny texts with similar lexical properties. The dataset comprises of more than 300,000 short texts, which is significantly larger than any previous humor-related corpus. Manual annotation of about 2,000 items proved the reliability of the corpus construction approach. Further, we applied language model fine-tuning for text classification and obtained an F1 score of 0.91 on test set, which constitutes a considerable gain over baseline methods. The dataset is freely available for research community.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…We thank volunteers who took part in data annotation. We are also thanksful to authors of (Ermilov et al, 2018), who ran their code on our data.…”

Section: Acknowledgmentsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Large Dataset and Language Model Fun-Tuning for Humor Recognition

Blinov¹,

Bolotova-Baranova

Braslavski³

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Self Cite

View full text Add to dashboard Cite

show abstract

Processing and Analysis of Russian Strategic Planning Programs

Alekseychuk

Sarkisyan

Emelyanov

et al. 2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

CLEF Workshop JOKER: Automatic Wordplay and Humour Translation

Ermakova

Miller

Puchalski

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Humour remains one of the most difficult aspects of intercultural communication: understanding humour often requires understanding implicit cultural references and/or double meanings, and this raises the question of its (un)translatability. Wordplay is a common source of humour in due to its attention-getting and subversive character. The translation of humour and wordplay is therefore in high demand. Modern translation depends heavily on technological aids, yet few works have treated the automation of humour and wordplay translation, or the creation of humour corpora. The goal of the JOKER workshop is to bring together translators and computer scientists to work on an evaluation framework for wordplay, including data and metric development, and to foster work on automatic methods for wordplay translation. We propose three pilot tasks: (1) classify and explain instances of wordplay, (2) translate single words containing wordplay, and (3) translate entire phrases containing wordplay.

show abstract

Stierlitz Meets SVM: Humor Detection in Russian

Cited by 12 publications

References 12 publications

Large Dataset and Language Model Fun-Tuning for Humor Recognition

Large Dataset and Language Model Fun-Tuning for Humor Recognition

Processing and Analysis of Russian Strategic Planning Programs

CLEF Workshop JOKER: Automatic Wordplay and Humour Translation

Contact Info

Product

Resources

About