A Supervised Method to Predict the Popularity of News Articles

Balali, Ali; Asadpour, Masoud; Faili, Hesham

doi:10.13053/cys-21-4-2848

Cited by 8 publications

(11 citation statements)

References 19 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Balali et al [2] took the number of comments as the base metric of the article's popularity. For building a reliable predictive model, authors extracted both textual (article title, body, comment tree) and non-textual (number of views, likes, dislikes) information from article web pages.…”

Section: Related Workmentioning

confidence: 99%

Prediction of News Popularity via Keywords Extraction and Trends Tracking

Pugachev

Voronov²,

Makarov

2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

In the last years, news agencies have become more influential in various social groups. At the same time, the media industry starts to monetize online distributed articles with contextual advertising. However, the efficiency of online marketing highly depends on the popularity of news articles. In our work, we present an alternative and effective way for article popularity forecasting with two-step approach: article keywords extraction and keywords-based article popularity prediction. We show the benefits of this technique and compare with widely used methods, such as Text Embeddings and BERT-based methods. Moreover, the work provides an architecture of the model for dynamic keyword tracking trained on the newest dataset of Russian news articles with more than 280k articles and 22k keywords for the popularity of forecasting purposes.

show abstract

Section: Related Workmentioning

confidence: 99%

Prediction of News Popularity via Keywords Extraction and Trends Tracking

Pugachev

Voronov²,

Makarov

2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…The data taken is all data uploaded up to February 6, 2019 for k-4 grades, January 3, 2019 for grades 5…”

Section: Data Collectionmentioning

confidence: 99%

“…According to data from the Pew Research center in 2005-2015, 65% of adults in the US use social media, where most traces are left in the form of comments about feelings, opinions, and ideas about individual, social, product, or even events that occur in around [5]. The distribution of data more and more every day raises the problem of the difficulty of monitoring the distribution of news or important events that occur and deserve to be used as literacy material [6].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Naive Bayes and Random Forests Classifier in the Classification of News Article Popularity as Learning Material

Pujianto¹,

Zaeni²,

Rasyida³

2021

Advances in Social Science, Education and Humanities Research

View full text Add to dashboard Cite

Tweentribune.com is a website that provides news texts with Lexile level information for each text. The Lexile level information feature is useful as a consideration for visitors to choose text that has a level of difficulty that matches their age. With such information, any news text on the site can be used as attractive teaching material for both teachers and students. However, the number of visits to each text varies widely. This study assumes that the popularity of each text is influenced not only by Lexile level information but also by other text characteristics. This research produces a number of engineering features that are extracted from the text to be used as a predictive attribute in classifying the popularity of the texts in question. Naïve Bayes and Random Forest are two classifiers used together with two popularity cluster scenarios based on k-means clustering. The results of the testing and evaluation phase show that the Random Forest algorithm has the best performance, with an accuracy value of 99.75%, an average recall value of 99.7%, and an average precision value of 98.7%.

show abstract

“…Some of the explored problems include examining the relationships between news comment topicality, temporality, sentiment, and quality (Diakopoulos and Naaman 2011; He et al 2020;; analyzing the sentiment of comments and headlines of news article (Dos Rieis et al 2015); news propagation (Tan, Friggeri, and Adamic 2016); personalized recommendation of news stories (Shmueli et al 2012); topic clustering of news articles (Aker et al 2016); and modeling and predicting comment volume (Tsagkias, Weerkamp, and De Rijke 2010;Balali, Asadpour, and Faili 2017;Rizos, Papadopoulos, and Kompatsiaris 2016;Tatar et al 2011). The prediction of comment volume is treated as a (binary) classification problem (e.g., "High"⁄ "Low" volume) (Tsagkias, Weerkamp, and De Rijke 2009) and regression classification problem (Balali, Asadpour, and Faili 2017) in previous studies. (Tatar et al 2011) uses a simple linear regression model with early user activity during a short observation period after publication to predict the comment volume of articles.…”

Section: Related Workmentioning

confidence: 99%

“…A key indicator of user participation in daily news events is the volume of user comments reacting to a news article (Prochazka, Weber, and Schweiger 2018;Ziegele et al 2018). Several works propose methods to predict it (Tsagkias, Weerkamp, and De Rijke 2009;Balali, Asadpour, and Faili 2017). They use a large number of features, which can be broadly categorized into article content, meta-article (e.g., outlet or category), temporal (e.g., date and time of publication), and semantic (e.g., named entities).…”

Section: Introductionmentioning

confidence: 99%

Cannot Predict Comment Volume of a News Article before (a few) Users Read It

Shen

Mukherjee

et al. 2020

Preprint

View full text Add to dashboard Cite

Many news outlets allow users to contribute comments on topics about daily world events. News articles are the seeds that spring users' interest to contribute content, i.e., comments. An article may attract an apathetic user engagement (several tens of comments) or a spontaneous fervent user engagement (thousands of comments). In this paper, we study the problem of predicting the total number of user comments a news article will receive. Our main insight is that the early dynamics of user comments contribute the most to an accurate prediction, while news article specific factors have surprisingly little influence. This appears to be an interesting and understudied phenomenon: collective social behavior at a news outlet shapes user response and may even downplay the content of an article. We compile and analyze a large number of features, both old and novel from literature. The features span a broad spectrum of facets including news article and comment contents, temporal dynamics, sentiment/linguistic features, and user behaviors. We show that the early arrival rate of comments is the best indicator of the eventual number of comments. We conduct an in-depth analysis of this feature across several dimensions, such as news outlets and news article categories. We show that the relationship between the early rate and the final number of comments as well as the prediction accuracy vary considerably across news outlets and news article categories (e.g., politics, sports, or health).

show abstract

A Supervised Method to Predict the Popularity of News Articles

Cited by 8 publications

References 19 publications

Prediction of News Popularity via Keywords Extraction and Trends Tracking

Prediction of News Popularity via Keywords Extraction and Trends Tracking

Comparison of Naive Bayes and Random Forests Classifier in the Classification of News Article Popularity as Learning Material

Cannot Predict Comment Volume of a News Article before (a few) Users Read It

Contact Info

Product

Resources

About