Stylometry Detection Using Deep Learning

K, Surendran; Harilal, O. P.; Hrudya, P.; Poornachandran, Prabaharan; Suchetha, N

doi:10.1007/978-981-10-3874-7_71

Cited by 12 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Author attribution via stylometry has traditionally focused on standard machine learning (ML) algorithms and feature engineering [1,22,33,56,68,74], but deep learning methods have become more prominent in recent years [4,9,19,67]. While there is no unanimous agreement on the most effective features [22,24,33], the Writeprints feature set has been widely applied with success [1-3, 15, 47, 53, 74].…”

Section: Introductionmentioning

confidence: 99%

Effective writing style transfer via combinatorial paraphrasing

Gröndahl

Asokan

2020

Proceedings on Privacy Enhancing Technologies

View full text Add to dashboard Cite

Stylometry can be used to profile or deanonymize authors against their will based on writing style. Style transfer provides a defence. Current techniques typically use either encoder-decoder architectures or rule-based algorithms. Crucially, style transfer must reliably retain original semantic content to be actually deployable. We conduct a multifaceted evaluation of three state-of-the-art encoder-decoder style transfer techniques, and show that all fail at semantic retainment. In particular, they do not produce appropriate paraphrases, but only retain original content in the trivial case of exactly reproducing the text. To mitigate this problem we propose ParChoice: a technique based on the combinatorial application of multiple paraphrasing algorithms. ParChoice strongly outperforms the encoder-decoder baselines in semantic retainment. Additionally, compared to baselines that achieve nonnegligible semantic retainment, ParChoice has superior style transfer performance. We also apply ParChoice to multi-author style imitation (not considered by prior work), where we achieve up to 75% imitation success among five authors. Furthermore, when compared to two state-of-the-art rule-based style transfer techniques, ParChoice has markedly better semantic retainment. Combining ParChoice with the best performing rulebased baseline (Mutant-X [34]) also reaches the highest style transfer success on the Brennan-Greenstadt and Extended-Brennan-Greenstadt corpora, with much less impact on original meaning than when using the rulebased baseline techniques alone. Finally, we highlight a critical problem that afflicts all current style transfer techniques: the adversary can use the same technique for thwarting style transfer via adversarial training. We show that adding randomness to style transfer helps to mitigate the effectiveness of adversarial training.

show abstract

Section: Introductionmentioning

confidence: 99%

Effective writing style transfer via combinatorial paraphrasing

Gröndahl

Asokan

2020

Proceedings on Privacy Enhancing Technologies

View full text Add to dashboard Cite

show abstract

“…In contrast to the previous study, [68] introduced 12 readability features to enhance accuracy. Together with three character-level features, five word-level features, one sentence richness feature and one vocabulary richness feature, with convolutional neural networks as the classifier, an accuracy of 97.7% and 90.1% was achieved for gender and age, respectively.…”

Section: ) Twittermentioning

confidence: 99%

“…On its own, the LIWC feature set yielded accuracy of 80.43% for the last 1000 tweets, which increased to 82.26% when augmented with the tweeting behavior. Character n-grams is another writing style technique seen in the following studies [34], [61], [67], [68]. [67] used every possible 1 through 5-gram for the 95 most-used ASCII characters to represent the tweet to infer gender.…”

Section: ) Twittermentioning

confidence: 99%

Online Social Networks and Writing Styles–A Review of the Multidisciplinary Literature

2020

View full text Add to dashboard Cite

In recent years, author identification has become an active research area, where the major differences are caused by paper or online medium, mode of entry and target audience. Much research has been devoted to analyzing writing styles in handwritten, word-processed and online social networks (OSN) texts. Word processing editors that typically include spell and grammar checkers may influence the writing style as it allows an individual to edit a piece of text to perfection. Thus, similarities may exist between OSN and word-processed texts. Moreover, none of the studies to date have made a detailed comparison of the writing styles across multidisciplinary factors. This paper attempts to close the gap between the writing styles in pre-and post-Internet periods as well as provide an in-depth comparison of the writing styles in OSN texts across three major factors: demographics, personality & behavior, and cybersecurity. The aim is to learn from past literature as we advance these techniques to OSN texts. Thus, in this paper, we also propose a novel machine learning prediction model based on tense morphology, to classify age and gender from English blogs, and the PAN 2013 dataset. This model achieves an accuracy of 94%-98% and 95%-97% for age and gender, respectively. INDEX TERMS Online social networks, survey, writing styles.

show abstract

“…SVMs have been particularly popular due to their strong performance on high-dimensional and sparse data [163]. Deep learning applications have recently become more prominent, with a particular focus on recurrent and convolutional neural networks [7,56,164]. Brocordo et al [16] also experiment with deep belief networks, which belong to the class of probabilistic generative models.…”

Section: Author Identificationmentioning

confidence: 99%

Text Analysis in Adversarial Settings

Gröndahl

Asokan

2019

ACM Comput. Surv.

View full text Add to dashboard Cite

Textual deception constitutes a major problem for online security. Many studies have argued that deceptiveness leaves traces in writing style, which could be detected using text classification techniques. By conducting an extensive literature review of existing empirical work, we demonstrate that while certain linguistic features have been indicative of deception in certain corpora, they fail to generalize across divergent semantic domains. We suggest that deceptiveness as such leaves no content-invariant stylistic trace , and textual similarity measures provide a superior means of classifying texts as potentially deceptive. Additionally, we discuss forms of deception beyond semantic content, focusing on hiding author identity by writing style obfuscation . Surveying the literature on both author identification and obfuscation techniques, we conclude that current style transformation methods fail to achieve reliable obfuscation while simultaneously ensuring semantic faithfulness to the original text. We propose that future work in style transformation should pay particular attention to disallowing semantically drastic changes.

show abstract

Stylometry Detection Using Deep Learning

Cited by 12 publications

References 9 publications

Effective writing style transfer via combinatorial paraphrasing

Effective writing style transfer via combinatorial paraphrasing

Online Social Networks and Writing Styles–A Review of the Multidisciplinary Literature

Text Analysis in Adversarial Settings

Contact Info

Product

Resources

About