Making computers laugh: investigations in automatic humor recognition
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Semi-supervised recognition of sarcastic sentences in Twitter and Amazon
CoNLL '10 Proceedings of the Fourteenth Conference on Computational Natural Language Learning
Classifying latent user attributes in twitter
SMUC '10 Proceedings of the 2nd international workshop on Search and mining user-generated contents
Democrats, republicans and starbucks afficionados: user classification in twitter
Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining
Hi-index | 0.00 |
Much has been written about humor and even sarcasm automatic recognition on Twitter. The task of classifying humorous tweets according to the type of humor has not been confronted so far, as far as we know. This research is aimed at applying classification and other NLP algorithms to the challenging task of automatically identifying the type and topic of humorous messages on Twitter. To achieve this goal, we will extend the related work surveyed hereinafter, adding different types of humor and characteristics to distinguish between them, including stylistic, syntactic, semantic and pragmatic ones. We will keep in mind the complex nature of the task at hand, which emanates from the informal language applied in tweets and variety of humor types and styles. These tend to be remarkably different from the type specific ones recognized in related works. We will use semi-supervised classifiers on a dataset of humorous tweets driven from different Twitter humor groups or funny tweet sites. Using a Mechanical Turk we will create a gold standard in which each tweet will be tagged by several annotators, in order to achieve an agreement between them, although the nature of the humor might allow one tweet to be classified under more than one class and topic of humor.