# Clickbait Convolutional Neural Network

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

- We proposed a clickbait convolutional neural network (CBCNN) model for the clickbait-detection problem. To the best of our knowledge, this is the first attempt to optimize a CNN model in clickbait detection.
- We designed a new word-embedding structure in this work. The new word-embedding layer takes both overall and type-related word meanings into consideration.
- We proposed a new loss function to regulate the influence of type-related word meaning.
- We conducted extensive experiments, and the results show that the CBCNN model outperforms all the five baseline methods in terms of accuracy, precision and recall.

## 2. Related Work

#### 2.1. Lexical Similarity Algorithms

#### 2.2. Machine Learning Algorithms

## 3. Methodology

#### 3.1. Word Embedding

#### 3.2. Clickbait Convolutional Neural Network

## 4. Experiments and Discussions

#### 4.1. Experiment Setup

- CIBTSS. Wang et al. [1] proposed a method detecting clickbait based on the lexical similarity between headline and content. Their method is named as CIBTSS.
- NB. The classic text classification method naive Bayes. We utilized unigrams and bigrams as the features of Bayes. The method that uses only unigrams is marked as $N{B}_{1gram}$, the method that uses only bigrams is named as $N{B}_{2gram}$, and the method that utilizes both unigrams and bigrams is marked as $N{B}_{1-2gram}$.
- PBCD. Biyani et al. [2] proposed series types of features for detecting clickbait, which is the latest machine learning-based method. Their features include unigram, bigram and a series of other newly defined features, such as the number of words, exclamatory marks and question marks. We named it as PBCD in this study.
- FastText. FastText [38] is a text-classification method similar to Word2Vec. Like Word2Vec, the sequence of words is considered in FastText. The learning algorithm of FastText is similar to the continuous bag-of-word (CBOW) model [39], which is a model learning distributed representations of words based on ordered words.
- TextCNN. TextCNN is evaluated by Agrawal [19]. As shown by Agrawal, the performance of TextCNN is the best among the five baselines.

#### 4.2. Experimental Results

#### 4.3. Discussion

- A number of clickbait articles tend to use similar words to attract users’ attention. Therefore, unigram-based machine learning algorithms figure out clickbait to a certain extent.
- Feature engineering is useful for clickbait detection, discarding the robustness problem.
- Word-sequence information helps machine learning algorithms to understand clickbait semantically.
- Various features are necessary for detecting clickbait, no matter if they had been extracted by feature engineering or by convolutional neural network means.
- The type-related features are important but undesirable when overvalued.

## 5. Conclusions and Future Work

## Author Contributions

## Acknowledgments

## Conflicts of Interest

## References

- Wang, Z.-C.; Weng, N.; Wang, Y. Research of Title Party News Identification Technology Based on Topic Sentence Similarity. New Technol. Lib. Inf. Serv.
**2011**, 11, 48–53. [Google Scholar] [CrossRef] - Biyani, P.; Tsioutsiouliklis, K.; Blackmer, J. “8 Amazing secrets for getting more clicks”: Detecting clickbait in news streams using article informality. In Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, Phoenix, AZ, USA, 12–17 February 2016. [Google Scholar]
- Chakraborty, A.; Paranjape, B.; Kakarla, S.; Ganguly, N. Stop clickbait: Detecting and preventing clickbait in online news media. In Proceedings of the IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, San Francisco, CA, USA, 18–21 August 2016; pp. 9–16. [Google Scholar]
- Potthast, M.; Köpsel, S.; Stein, B.; Hagen, M. Clickbait Detection. In Proceedings of the 38 European Conference on Information Retrieval, Padua, Italy, 20–23 March 2016. [Google Scholar]
- Chen, Y.; Conroy, N.J.; Rubin, V.L. Misleading online content: Recognizing clickbait as “False News”. In Proceedings of the ACM Workshop on Multimodal Deception Detection, Seattle, WA, USA, 9–13 November 2015; pp. 15–19. [Google Scholar]
- Abbasi, A.; Zhang, Z.; Zimbra, D.; Chen, H.; Nunamaker, J.F. Detecting fake websites: The contribution of statistical learning theory. Mis Q.
**2010**, 34, 435–461. [Google Scholar] [CrossRef] - Abbasi, A.; Chen, H. A comparison of fraud cues and classification methods for fake escrow website detection. Inf. Technol. Manag.
**2009**, 10, 83–101. [Google Scholar] [CrossRef] - Ntoulas, A.; Najork, M.; Manasse, M.; Fetterly, D. Detecting spam web pages through content analysis. In Proceedings of the World Wide Web Conference, Edinburgh, Scotland, 23–26 May 2006; pp. 83–92. [Google Scholar]
- Lahiri, S.; Mitra, P.; Lu, X. Informality judgment at sentence level and experiments with formality score. In CICLing 2011: Computational Linguistics and Intelligent Text Processing; Springer: Berlin/Heidelberg, Germany, 2011; pp. 446–457. [Google Scholar]
- Krizhevsky, A.; Sutskever, I.; Hinton, G.E. ImageNet classification with deep convolutional neural networks. In Proceedings of the 25th International Conference on Neural Information Processing Systems, Lake Tahoe, NV, USA, 3–6 December 2012; pp. 1097–1105. [Google Scholar]
- Haykin, S.; Kosko, B. GradientBased Learning Applied to Document Recognition; IEEE: New York, NY, USA, 2009; pp. 306–351. [Google Scholar]
- Kim, Y. Convolutional Neural Networks for Sentence Classification. Eprint Arxiv, 2014; arXiv:1408.5882. [Google Scholar]
- Mikolov, T.; Le, Q.V.; Sutskever, I. Exploiting Similarities among Languages for Machine Translation. Comput. Sci. 2013. [Google Scholar] [CrossRef]
- Le, Q.V.; Mikolov, T. Distributed Representations of Sentences and Documents. Comput. Sci.
**2014**, 4, 1188–1196. [Google Scholar] - Collobert, R.; Weston, J.; Bottou, L.; Karlen, M.; Kavukcuoglu, K.; Kuksa, P. Natural language processing (almost) from scratch. J. Mach. Learn. Res.
**2011**, 12, 2493–2537. [Google Scholar] - Zeng, D.; Liu, K.; Chen, Y.; Zhao, J. Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, Edinburgh, UK, 27–31 July 2015; pp. 1753–1762. [Google Scholar]
- Chen, Y.; Xu, L.; Liu, K.; Zeng, D.; Zhao, J. Event Extraction via Dynamic Multi-Pooling Convolutional Neural Networks. In Proceedings of the Association for Computational Linguistics, Beijing, China, 26–31 July 2015. [Google Scholar]
- He, H.; Gimpel, K.; Lin, J. Multi-Perspective Sentence Similarity Modeling with Convolutional Neural Networks. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, Lisbon, Portuga, 17–21 September 2015; pp. 1576–1586. [Google Scholar]
- Agrawal, A. Clickbait detection using deep learning. In Proceedings of the IEEE 2016 2nd International Conference on Next Generation Computing Technologies (NGCT), Dehradun, India, 14–16 October 2016; pp. 268–272. [Google Scholar]
- Gabrilovich, E.; Markovitch, S. Computing semantic relatedness using Wikipedia-based explicit semantic analysis. In Proceedings of the 20th International Joint Conference on Artificial Intelligence, Hyderabad, India, 6–12 January 2007; pp. 1606–1611. [Google Scholar]
- Kiros, R.; Zhu, Y.; Salakhutdinov, R.; Zemel, R.S.; Torralba, A.; Urtasun, R.; Fidler, S. Skip-Thought Vectors. Comput. Sci. 2015; arXiv:1506.06726. [Google Scholar]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V. Scikit-learn: Machine Learning in Python. J. Mach. Learn. Res.
**2012**, 12, 2825–2830. [Google Scholar] - Friedman, J.H. Greedy Function Approximation: A Gradient Boosting Machine. Ann. Stat.
**2001**, 29, 1189–1232. [Google Scholar] [CrossRef] - Cortes, C.; Vapnik, V. Support-vector networks. Mach. Learn.
**1995**, 20, 273–297. [Google Scholar] [CrossRef] - John, G.H.; Langley, P. Estimating Continuous Distributions in Bayesian Classifiers. In Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, Montréal, QC, Canada, 18–20 August 2013; pp. 338–345. [Google Scholar]
- Molek-Kozakowska, K. Coercive Metaphors in News Headlines a Cognitive-Pragmatic Approach. Brno Stud. Engl.
**2014**, 40, 149–173. [Google Scholar] [CrossRef] - Cessie, S.L.; Houwelingen, J.C.V. Ridge Estimators in Logistic Regression. J. R. Stat. Soc.
**1992**, 41, 191–201. [Google Scholar] [CrossRef] - Lewis, D.D. Naive (Bayes) at Forty: The Independence Assumption in Information Retrieval; Springer: Berlin/Heidelberg, Germany, 1998; pp. 1204–1206. [Google Scholar]
- Breiman, L. Random Forests. Mach. Learn.
**2001**, 45, 5–32. [Google Scholar] [CrossRef] - Zheng, H.T.; Yao, X.; Jiang, Y.; Xia, S.T.; Xiao, X. Boost clickbait detection based on user behavior analysis. In APWeb-WAIM 2017: Web and Big Data; Springer: Cham, Switzerland, 2017; pp. 73–80. [Google Scholar]
- Anand, A.; Chakraborty, T.; Park, N. We used Neural Networks to detect clickbait: You won’t believe what happened Next! In ECIR 2017: Advances in Information Retrieval; Springer: Cham, Switzerland, 2017; pp. 541–547. [Google Scholar]
- Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput.
**1997**, 9, 1735–1780. [Google Scholar] [CrossRef] [PubMed] - Cho, K.; Van Merriënboer, B.; Bahdanau, D.; Bengio, Y. On the properties of neural machine translation: Encoder-decoder approaches. arXiv, 2014; arXiv:1409.1259. [Google Scholar]
- Santos, C.D.; Zadrozny, B. Learning character-level representations for part-of-speech tagging. In Proceedings of the 31st International Conference on Machine Learning (ICML-14), Beijing, China, 21–26 June 2014; pp. 1818–1826. [Google Scholar]
- Janocha, K.; Czarnecki, W.M. On Loss Functions for Deep Neural Networks in Classification. Schedae Inform.
**2016**, 25, 49–59. [Google Scholar] [CrossRef] - Hinton, G.E.; Srivastava, N.; Krizhevsky, A.; Sutskever, I.; Salakhutdinov, R.R. Improving neural networks by preventing co-adaptation of feature detectors. Comput. Sci.
**2012**, 3, 212–223. [Google Scholar] - Bouvrie, J.; Massachusetts Institute of Technology, Cambridge, MA, USA. Notes on Convolutional Neural Networks. Unpublished work. 2006. [Google Scholar]
- Joulin, A.; Grave, E.; Bojanowski, P.; Mikolov, T. Bag of Tricks for Efficient Text Classification. Comput. Sci. 2016; arXiv:1607.01759. [Google Scholar]
- Mikolov, T.; Chen, K.; Corrado, G.; Dean, J. Efficient Estimation of Word Representations in Vector Space. Comput. Sci. 2013; arXiv:1301.3781. [Google Scholar]

$\mathit{\alpha}=0.25$ | $\mathit{\alpha}=0.5$ | $\mathit{\alpha}=1$ | $\mathit{\alpha}=2$ | $\mathit{\alpha}>=4$ | TextCNN | |
---|---|---|---|---|---|---|

Precision | 69.21% | 71.32% | 73.37% | 72.59% | 71.74% | 71.74% |

Recall | 81.34% | 84.31% | 88.21% | 87.35% | 86.48% | 86.48% |

Accuracy | 74.16% | 77.54% | 80.5% | 79.84% | 78.18% | 78.18% |

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Zheng, H.-T.; Chen, J.-Y.; Yao, X.; Sangaiah, A.K.; Jiang, Y.; Zhao, C.-Z.
Clickbait Convolutional Neural Network. *Symmetry* **2018**, *10*, 138.
https://doi.org/10.3390/sym10050138

**AMA Style**

Zheng H-T, Chen J-Y, Yao X, Sangaiah AK, Jiang Y, Zhao C-Z.
Clickbait Convolutional Neural Network. *Symmetry*. 2018; 10(5):138.
https://doi.org/10.3390/sym10050138

**Chicago/Turabian Style**

Zheng, Hai-Tao, Jin-Yuan Chen, Xin Yao, Arun Kumar Sangaiah, Yong Jiang, and Cong-Zhi Zhao.
2018. "Clickbait Convolutional Neural Network" *Symmetry* 10, no. 5: 138.
https://doi.org/10.3390/sym10050138