Embedding Lexical Features via Tensor Decomposition for Small Sample Humor Recognition

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 6376–6381, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 6376 Embedding Lexical Features via Tensor Decomposition for Small Sample Humor Recognition Zhenjie Zhao1∗ Andrew Cattle1∗ Evangelos E. Papalexakis2† Xiaojuan Ma1 1 Department of Computer Science & Engineering, HKUST 2 Department of Computer Science & Engineering, University of California, Riverside {zzhaoao, acattle}@connect.ust.hk, epapalex@cs.ucr.edu, mxj@cse.ust.hk Abstract We propose a novel tensor embedding method that can effectively extract lexical features for humor recognition. Specifically, we use word- word co-occurrence to encode the contextual content of documents, and then decompose the tensor to get corresponding vector repre- sentations. We show that this simple method can capture features of lexical humor effec- tively for continuous humor recognition. In particular, we achieve a distance of 0.887 on a global humor ranking task, comparable to the top performing systems from SemEval 2017 Task 6B (Potash et al., 2017) but without the need for any external training corpus. In ad- dition, we further show that this approach is also beneficial for small sample humor recog- nition tasks through a semi-supervised label propagation procedure, which achieves about 0.7 accuracy on the 16000 One-Liners (Mi- halcea and Strapparava, 2005) and Pun of the Day (Yang et al., 2015) humour classification datasets using only 10% of known labels. 1 Introduction Recognizing humor automatically is an impor- tant step for natural human-computer interaction (Shahaf et al., 2015). While early works tend to frame humor recognition as a binary classifica- tion task (Mihalcea and Strapparava, 2005; Yang et al., 2015), the last few years have seen the emer- gence of humor recognition as a pairwise relative ranking task (Cattle and Ma, 2016; Shahaf et al., 2015). In addition to pairwise ranking, SemEval 2017 Task 6 also includes a global ranking sub- task. However, the majority of submissions build ∗Zhenjie Zhao and Andrew Cattle contributed equally to this work. † E. Papalexakis was supported by a UCR-China collab- oration grant by the Bourns College of Engineering at UCR, and by the National Science Foundation CDSE Grant no. OAC-1808591 global rankings using a series of pairwise compar- isons (Potash et al., 2017). Only Yan and Pedersen (2017) attempt to predict global rankings directly, ranking documents inversely to their probability according to an n-gram language model. State-of-the-art humor recognition algorithms usually require a considerable amount of training data with labels to learn effective features (Yang et al., 2015). However, such data are difficult to obtain – especially fine-grained humor anno- tations. First, the humor judgments differ from individual to individual. Thus, collecting percep- tually consistent human labels is expensive and time-consuming. Second, fine-grained degrees of humor add a further challenge. Therefore, meth- ods on small sample learning or even unsupervised rule-based methods merit investigation. In this paper, considering the importance of lexical information for humor recognition (Radev et al., 2015), we propose a tensor decomposition method to capture the contextual …

Content truncated to 500 words