Personalizing news content: An experimental study

Abstract

The delivery of personalized news content depends on the ability to predict user interests. We evaluated different methods for acquiring user profiles based on declared and actual interest in various news topics and items. In an experiment, 36 students rated their interest in six news topics and in specific news items and read on 6 days standard, nonpersonalized editions and personalized (basic or adaptive) news editions. We measured subjective satisfaction with the editions and expressed preferences, along with objective measures, to infer actual interest in items. Users' declared interest in news topics did not strongly predict their actual interest in specific news items. Satisfaction with all news editions was high, but participants preferred the personalized editions. User interest was weakly correlated with reading duration, article length, and reading order. Different measures predicted interest in different news topics. Explicit measures predicted interest in relatively clearly defined topics such as sports, but were less appropriate for broader topics such as science and technology. Our results indicate that explicit and implicit methods should be combined to generate user profiles. We suggest that a personalized newspaper should contain both general information and personalized items, selected based on specific combinations of measures for each of the different news topics. Based on the findings, we present a general model to decide on the personalization of news content to generate personalized editions for readers.

The personal information sphere: An integral approach to privacy and related information and communication rights

“…Furthermore, an experiment by Eslami et al (2015) suggests that users become more engaged and feel they have control when they are made aware of the personalization algorithms. At the same time, the topics that people say they are interested in do not always match with the topics that their clicking behavior shows they are actually interested in Sela, Lavie, Inbar, Oppenheim, and Meyer (2015). Full and unlimited user control might thus not be in the user's interest.…”

Effects of Uniqueness, News Valence, and Liking on Personalization of Company News

“…How to balance the need to provide news consumers with what they want to read (e.g., news stories consistent with their preexisting beliefs) and what they need to read (e.g., news stories challenging their preexisting beliefs) is a critical issue to consider in building effective and beneficial news recommendation systems. As an example, Sela et al (2015) argued that online news systems should include both personalized news stories and general news stories. Although users may not explicitly express preferences for general and standardized news stories, those stories may still interest them and be beneficial to read.…”

“…Algorithmic recommendation systems are being widely used by search engines and social media platforms today to personalize news for users. People in general prefer personalized news stories in alignment with their preferences than standardized news articles picked by traditional media gatekeepers, but the main challenge of presenting such personalized news content to them is the ability to predict their particular news interests (Sela et al, 2015). If users’ preferences over different news content areas can be accurately assessed and modeled, it can help guide newspaper space allocation to enhance circulation and revenue (Kanuri et al, 2014).…”

Capturing the change in topical interests of personas over time

“…Personalizing the delivery of news and other online content depends on the ability to predict user interests, with tailored content being preferred by users, although declared interest and actual interest may differ (Sela, Lavie, Inbar, Oppenheim, & Meyer, 2015). Therefore, developing personalized content systems requires not only detecting and tracking current interests but also predicting what content users would be interested in the future (Mele, Bahrainian, & Crestani, 2019) or at least what related future content (Toraman & Can, 2017) a person might prefer.…”