An attention network via pronunciation, lexicon and syntax for humor recognition

Ren, Lu; Xu, Bo; Lin, Hongfei; Zhang, Jinhui; Yang, Liang

doi:10.1007/s10489-021-02580-3

Cited by 8 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhang [13] interactively modeled the semantic information of setups and punchlines in humorous texts at both word and clause levels to predict the humor level. Ren [14] used CNN, LSTM [15] and Attention to encode the text from pronunciation, lexicon and syntax perspectives to recognize humor.…”

Section: Related Workmentioning

confidence: 99%

Combining character information and contextual feature for conversational humor recognition

Zhao,

Zhang

2023

Third International Conference on Artificial Intelligence, Virtual Reality, and Visualization (AIVRV 2023)

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Combining character information and contextual feature for conversational humor recognition

Zhao,

Zhang

2023

Third International Conference on Artificial Intelligence, Virtual Reality, and Visualization (AIVRV 2023)

View full text Add to dashboard Cite

“…In recent years, Deep Neural Networks (DNNs) have been employed. To give an example, Chen and Soo [28] build a Convolutional Neural Network (CNN), while Ren et al [35] propose an approach based on Long Short-Term Memorys (LSTMs) and an attention mechanism. With the advent of large pretrained transformer language models like BERT [36], the focus of textual humour detection shifted towards such models, motivated by their promising performance in many NLP downstream tasks.…”

Section: Textual Humour Recognitionmentioning

confidence: 99%

Multimodal Prediction of Spontaneous Humour: A Novel Dataset and First Results

Christ¹,

Amiriparian²,

Kathan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Humour is a substantial element of human affect and cognition. Its automatic understanding can facilitate a more naturalistic human-device interaction and the humanisation of artificial intelligence. Current methods of humour detection are solely based on staged data making them inadequate for 'real-world' applications. We address this deficiency by introducing the novel Passau-Spontaneous Football Coach Humour (Passau-SFCH) dataset, comprising of about 11 hours of recordings. The Passau-SFCH dataset is annotated for the presence of humour and its dimensions (sentiment and direction) as proposed in Martin's Humor Style Questionnaire. We conduct a series of experiments, employing pretrained Transformers, convolutional neural networks, and expert-designed features. The performance of each modality (text, audio, video) for spontaneous humour recognition is analysed and their complementarity is investigated. Our findings suggest that for the automatic analysis of humour and its sentiment, facial expressions are most promising, while humour direction can be best modelled via text-based features. The results reveal considerable differences among various subjects, highlighting the individuality of humour usage and style. Further, we observe that a decision-level fusion yields the best recognition result. Finally, we make our code publicly available at https://www.github.com/EIHW/passau-sfch. The Passau-SFCH dataset is available upon request.

show abstract

“…In their study, Ren et al (2022) addressed jokes recognition by focusing on the linguistics elements. This study pointed out the attention to the aspects of pronunciation, lexicon, and syntax for jokes recognition.…”

Section: Introductionmentioning

confidence: 99%

“Hi, Google! Tell Me A Joke”: How Google Assistant Creates Its Jokes

Kurniawati¹,

Harared²,

Rohardiyanto³

2022

Metathesis.J.Engl.Lang.Lit.Teach.

View full text Add to dashboard Cite

Artificial intelligent-based application helps people in many ways, one of which telling them jokes. This study aims at (1) investigating the types of jokes on Google Assistant application, and (2) describing linguistic aspects used in the jokes. There were several steps in conducting this study. Firstly, the data of this study were the jokes taken from “tell me a joke” feature on Google Assistant. As many as 58 jokes were collected as the data. Then, the data were coded and analysed based on its linguistic aspects in terms of sounds, meanings, and word formations. After all data were analysed, conclusions were drawn. There are two forms of jokes in this application namely riddle and one liner. In riddle, the jokes are presented in question-answer form, whereas one liner joke only consists of one or two sentences. There are several linguistic aspects in creating the jokes, namely homonyms, minimal pair, substitution of sounds, onomatopoeia, phrase and words with similar pronunciation, idiomatic phrase, addition of sound and word combination. Among those aspects, homonyms are the most frequently utilized in the jokes. Jokes are beneficial for language learners to learn the real use of the language, thus further studies on jokes are suggested.

show abstract

An attention network via pronunciation, lexicon and syntax for humor recognition

Cited by 8 publications

References 32 publications

Combining character information and contextual feature for conversational humor recognition

Combining character information and contextual feature for conversational humor recognition

Multimodal Prediction of Spontaneous Humour: A Novel Dataset and First Results

“Hi, Google! Tell Me A Joke”: How Google Assistant Creates Its Jokes

Contact Info

Product

Resources

About