OFAI-UKP at HAHA@IberLEF2019: Predicting the Humorousness of Tweets Using Gaussian Process Preference Learning

@article{Miller2019OFAIUKPAH,
  title={OFAI-UKP at HAHA@IberLEF2019: Predicting the Humorousness of Tweets Using Gaussian Process Preference Learning},
  author={Tristan Miller and E. Dinh and Edwin Simpson and Iryna Gurevych},
  journal={Proces. del Leng. Natural},
  year={2019},
  volume={64},
  pages={37-44}
}
This work has been supported by the German Federal Ministry of Education and Research (BMBF) under the promotional reference 01UG1816B (CEDIFOR), by the German Research Foundation (DFG) as part of the QA-EduInf project (grants GU798/18-1 and RI 803/12-1), by the DFG-funded research training group “Adaptive Preparation of Information from Heterogeneous Sources” (AIPHES; GRK1994/1), and by the Austrian Science Fund (FWF) under project M2625-N31. The Austrian Research Institute for Artificial… Expand

Figures, Tables, and Topics from this paper

ELMo-NB at SemEval-2020 Task 7: Assessing Sense of Humor in EditedNews Headlines Using ELMo and NB
Our approach is constructed to improve on a couple of aspects; preprocessing with an emphasis on humor sense detection, using embeddings from state-of-the-art language model(Elmo), and ensembling theExpand
Overview of HAHA at IberLEF 2019: Humor Analysis based on Human Annotation
TLDR
This paper presents the results of the HAHA task at IberLEF 2019, the second edition of the challenge on automatic humor recognition and analysis in Spanish, and presents a summary of their systems and the general results for both subtasks. Expand
SSN_NLP at SemEval-2020 Task 7: Detecting Funniness Level Using Traditional Learning with Sentence Embeddings
Assessing the funniness of edited news headlines task deals with estimating the humorness in the headlines edited with micro-edits. This task has two sub-tasks in which one has to calculate the meanExpand

References

SHOWING 1-10 OF 52 REFERENCES
Overview of the HAHA Task: Humor Analysis Based on Human Annotation at IberEval 2018
This paper presents the HAHA task at IberEval 2018, the first challenge on humor appreciation in the Spanish language. The challenge proposes two tasks related to humor in language: automaticExpand
UO_UPV: Deep Linguistic Humor Detection in Spanish Social Media
TLDR
The UO UPV system developed for the Humor Analysis based on Human Annotation track proposed in IberEval 2018 Workshop, which combines both linguistic features and an Attention-based Recurrent Neural Network, where the attention layer helps to calculate the contribution of each term towards targeted humorous classes. Expand
Overview of HAHA at IberLEF 2019: Humor Analysis based on Human Annotation
TLDR
This paper presents the results of the HAHA task at IberLEF 2019, the second edition of the challenge on automatic humor recognition and analysis in Spanish, and presents a summary of their systems and the general results for both subtasks. Expand
Predicting and Characterising User Impact on Twitter
TLDR
This paper addresses the task of predicting the impact of Twitter users based only on features under their direct control, such as usage statistics and the text posted in their tweets, and applies linear as well as non-linear learning methods to predict a user impact score. Expand
A Crowd-Annotated Spanish Corpus for Humor Analysis
TLDR
A corpus of 27,000 tweets written in Spanish and crowd-annotated by their humor value and funniness score, with about four annotations per tweet, tagged by 1,300 people over the Internet is presented. Expand
Predicting Humorousness and Metaphor Novelty with Gaussian Process Preference Learning
TLDR
A Bayesian approach for predicting humorousness and metaphor novelty using Gaussian process preference learning (GPPL), which achieves a Spearman’s ρ of 0.56 against gold using word embeddings and linguistic features is introduced. Expand
Language Understanding in the Wild: Combining Crowdsourcing and Machine Learning
TLDR
This work presents a novel Bayesian approach to language understanding that relies on aggregated crowdsourced judgements, and encodes the relationships between labels and text features in documents, such as tweets, web articles, and blog posts, accounting for the varying reliability of human labellers. Expand
Leveraging Large Amounts of Weakly Supervised Data for Multi-Language Sentiment Classification
TLDR
This paper presents a novel approach for multi-lingual sentiment classification in short texts by leveraging large amounts of weakly-supervised data in various languages to train a multi-layer convolutional network and demonstrates the importance of using pre-training of such networks. Expand
Capturing Reliable Fine-Grained Sentiment Associations by Crowdsourcing and Best–Worst Scaling
TLDR
It is shown that on all three domains the ranking of words by sentiment remains remarkably consistent even when the annotation process is repeated with a different set of annotators, and the minimum difference in sentiment association that is perceptible to native speakers of a language is determined. Expand
Humor Analysis Based on Human Annotation Challenge at IberLEF 2019: First-place Solution
TLDR
The winning solution to the Humor Analysis based on Human Annotation (HAHA) task at IberLEF 2019 is described, using an ensemble of a fine-tuned multilingual BERT model and a naive Bayes classifier. Expand
...
1
2
3
4
5
...