"8 Amazing Secrets for Getting More Clicks": Detecting Clickbaits in News Streams Using Article Informality

Abstract

Clickbaits are articles with misleading titles, exaggerating the content on the landing page. Their goal is to entice users to click on the title in order to monetize the landing page. The content on the landing page is usually of low quality. Their presence in user homepage stream of news aggregator sites (e.g., Yahoo news, Google news) may adversely impact user experience. Hence, it is important to identify and demote or block them on homepages. In this paper, we present a machine-learning model to detect clickbaits. We use a variety of features and show that the degree of informality of a webpage (as measured by different metrics) is a strong indicator of it being a clickbait. We conduct extensive experiments to evaluate our approach and analyze properties of clickbait and non-clickbait articles. Our model achieves high performance (74.9% F-1 score) in predicting clickbaits.

Click me…! The influence of clickbait on user engagement in social media and the role of digital nudging

“…The strongest effect was found for the share count. This result contrasts with literature that proposed a positive effect of longer texts for headlines [ 52 , 57 , 58 ] which led to the hypothesis. However, the analysis showed that doubling the number of headline words (log-transformed with base 2) led to 23.7% fewer comments; the effects on reactions and shares were not statistically significant.…”

“…Regarding the use of questions, the present results indicate a negative effect on the interaction for headlines and post texts, however 50% of the results were not significant, prohibiting a solid hypothesis confirmation. As Biyani et al (2016) [ 57 ] found questions were an important clickbait feature, we also expected questions to evoke interaction, and in particular, comments. Kuiken et al (2017) [ 52 ] hypothesized a positive effect of questions on the click-through rate as well but found the opposite effect for article headlines.…”

“…To assess the presence of clickbait phrases, we used the list from the Downworthy browser plugin [ 76 ]. We used a top nine list containing clickbait unigrams of headlines for the unigram feature [ 57 ]. The latter two features as well as the punctuation feature were treated as Boolean variables (present or non-present), and the other three as numerical variables.…”

“…They comprise more questions as they should activate the reader [ 52 ]. Certain words or phrases in headlines (e.g., “you won’t believe”) have been identified as another indicator of clickbait [ 52 , 57 ]. Stop words and hyperbolic words that express highly positive sentiments (e.g., “best ever”) are frequently used [ 58 ].…”

“…Informed by studies that aimed either to classify articles into clickbait and non-clickbait [ 57 ] or measured the effectiveness of specific clickbait strategies for the click-through rate [ 52 ], we hypothesize that clickbait features in post texts as well as preview headlines influence (a) the number of reactions, (b) the number of shares and (c) the number of comments. The included features and their hypothesized direction are depicted in Table 1 .…”