With its linguistic, social, and psychological components, humor has long been a part of human life. Allowing computers to interpret humor has become extremely important due to its wide range of uses and growing popularity on social media platforms. While humor has long been researched from psychological, cognitive, and linguistic perspectives, computational linguistics has yet to investigate it. Our contribution to HAHA@IberLEF2021: Humor Analysis based on Human Annotation is described in this paper. We o er a deep neural network-based approach in this paper. Our research team uses multi-kernel convolution recurrent neural network model for the humor detection in tweets. We examine our method's performance and show how each component of our design contributes to its overall success.
Humor is a universal and subtle emotion that can be found all over the world.
The majority of prior research and studies on humor di culties were focused
on binary categorization or the identi cation of linguistic features. Purandare
and Litman utilized typical supervised learning classi ers to recognize hilarious
speech in a hilarious spoken dialogue as data from a famous comedy television
show [18]. Taylor and Mazlack used the methodology that was based on the
extraction of structural patterns and peculiar structure of jokes newcite [20].
Luke de Oliveira and Lainez applied recurrent neural network (RNN) and
convolutional neural networks (CNNs) to humor detection from reviews in Yelp
dataset [
Because tweets are brief and informal user-generated text that typically do
not follow grammatical standards, detecting humor in them presents particular
problems to the research community. Furthermore, tweets feature a plethora of
unique abbreviations as well as Twitter-speci c syntaxes like hashtags and
emojis. To address the challenge of humor detection, Chiruzzo et al. [
The rest of the paper is structured as follows: Section 2 presents a summary of previous studies. In Section 3, we introduce our proposed humor detection framework. Section 4 includes experiments and evaluations. Some concluded remarks of our work are described in Section 5. 2
In the related work of humor identi cation, there are a lot of work that is done
over the year which includes statistical and N-gram analysis [20], Regression
Trees [18], Word2Vec combined with K-NN Human Centric Features, and
Convolutional Neural Networks [
In this section, we describe the details of our proposed framework for humor detection is Spanish tweets. Figure 1 depicts an overview of our proposed framework.
We utilize the pre-trained-word-vectors-for-spanish word embedding purposes. The embedding matrix is fed into the embedding layer of our neural network. We start by extracting higher-level feature sequences from the target added tweet embeddings using multi-kernel convolution lters. These feature sequences are supplied into the Bi-LSTM that is linked to it. Following that, we go through each component in detail.
Multi-kernel CNN
kernel size [
Pooling Layer Bi-LSTM Layer
Dimensions = 300,
Vectors=1,000,653 Feature Vector
Forward Layer
Backward Layer
Starting with random weights, the Embedding layer in Figure 1 will learn an embedding for all of the words in the training dataset. The initial hidden layer of a network is de ned as this exible layer. A pre-trained model used in the embedding layer simply required a le containing tokens and their associated word vectors. The pre-trained word vectors for Spanish model was built using 300-dimensional word vectors. Dimensionality is a term that refers to how many dimensions there are in anything. The Embedding matrix will have a dimensionality of L x D, where L is the sentence length and D is the word-vector dimension. 3.2
Convolution Neural System We use the technique given by [13] to extract higher-level features in our multikernel convolution. The embedding matrix created in the embedding layer is the module's input. Then, using a lter, we apply convolution on it. We use three distinct kernel sizes, or the size of the convolution lters, to apply multiple convolutions: 3, 4, and 5. Each lter creates the matching feature maps after performing convolutions, and then a max-pooling function is used to build a univariate feature vector. Finally, each kernel's feature vectors are concatenated to create a single high-level feature vector. 3.3
Bi-LSTM Bidirectional Long Short Term Memory (Bi-LSTM) is a bidirectional variant of LSTM seen in the center of Figure 1. Bi-LSTM combines the forward and backward hidden layers, allowing access to both the previous and subsequent contexts. The Bi-LSTM neural network is used to obtain a vector representation of the input sentence that captures the semantics of the phrase e ectively. The nal result from Bi-LSTM's output layer is formed by merging the results from both RNN hidden layers, namely the forward and backward layers. 3.4
Humor classi cation and Prediction We get our results from the last linear layer of the model. We consider binary cross-entropy and mean square error (mse) as the loss function in sub-task1 and sub-task2, respectively. We use the stochastic gradient descent (SGD) to learn the model parameter and adopt the Adam optimizer [14].
Dataset The organizer provide a corpus of crowd-annotated tweets divided into three subsets for tasks 1 and 2: training (24,000 tweets), development (4,000 tweets), and testing (6,000 tweets). The annotation has a voting system in which users can choose from six di erent choices. The tweet is either not funny or funny, with a score ranging from one (not funny) to ve (very hilarious) (excellent).
To prepare the data, we eliminated stop words using NLTK's standard
stoplist, eliminated special characters, and performed hashtag segmentation using
the hashtag segmentation tool [
Model Con guration
In the following, we describe the set of parameters that we have used in our
framework during experiments. We used one embedding model to initialize the
word embeddings in the embedding layer. The embedding model has 300-dimensional
with 1,000,653 vectors. It is trained on Spanish Billion Word Corpus which has
the size of 1.4 billion words [
Results and Analysis Our target is to classify the tweets into humorous or not humorous from Spanish tweets. In Table 1, at rst we reported the results of sub-task1 and sub-task2 based on a Naive Bayes with t df features and SVM regression with t df features, respectively. Next, we reported the results of our proposed framework that were submitted in the competition. After the competition, some parameters were changed to check the betterment of results, and we found that we could improve our results so far to 77.453 percent F1 for sub-task1, and 0.6977 RMSE error for sub-task2, although these results could not be submitted to the competition. 5
Our technique for HAHA@IberLEF2021: Humor Analysis based on Human Annotation Forum was described in this article. Humor detection is a di cult process. We used deep learning techniques to try to solve the problem. We ran some tests with other models, such as a basic regression model and a multilayer perceptron model, however the model described in this paper was the one that produced the best results. In a summary, our uni ed framework's key contribution is that it successfully learns contextual information, which improves comedy detection performance. We want to leverage external data to generalize our model for comedy identi cation in the same region in the future.
This research was supported by the Japan International Cooperation Agency { JICA under Innovative Asia program. 11. Gupta, A., Pal, A., Khurana, B., Tyagi, L., Modi, A.: Humor@iitk at semeval-2021 task 7: Large language models for quantifying humor and o ensiveness (04 2021) 12. Kiddon, C., Brun, Y.: That's what she said: double entendre identi cation. In: Proceedings of the 49th annual meeting of the association for computational linguistics: Human language technologies. pp. 89{94 (2011) 13. Kim, Y.: Convolutional Neural Networks for Sentence Classi cation pp. 1746{1751 (2014) 14. Kingma, D.P., Ba, J.L.: A : a m s o pp. 1{15 (2015) 15. Mihalcea, R., Strapparava, C.: Making computers laugh: Investigations in automatic humor recognition. In: Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing. pp. 531{538 (2005) 16. Owens, J.D., Houston, M., Luebke, D., Green, S., Stone, J.E., Phillips, J.C.: Gpu computing. Proceedings of the IEEE 96(5), 879{899 (2008) 17. Potash, P., Romanov, A., Rumshisky, A.: SemEval-2017 task 6: #HashtagWars: Learning a sense of humor. In: Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017). pp. 49{57. Association for Computational Linguistics, Vancouver, Canada (Aug 2017). https://doi.org/10.18653/v1/S172004, bluehttps://www.aclweb.org/anthology/S17-2004 18. Purandare, A., Litman, D.: Humor: Prosody analysis and automatic recognition for f* r* i* e* n* d* s. In: Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing. pp. 208{215 (2006) 19. Ritchie, G.: Developing the incongruity-resolution theory. Tech. rep. (1999) 20. Taylor, J.M., Mazlack, L.J.: Computationally Recognizing Wordplay in Jokes Theories of Humor (1991) (2000) 21. Taylor, J.M., Mazlack, L.J.: Computationally recognizing wordplay in jokes. In: Proceedings of the Annual Meeting of the Cognitive Science Society. vol. 26 (2004)