Terms of a feather: content-based news recommendation and discovery using twitter
read more
Citations
DRN: A Deep Reinforcement Learning Framework for News Recommendation
A Survey of Social-Based Routing in Delay Tolerant Networks: Positive and Negative Social Effects
NPA: Neural News Recommendation with Personalized Attention
Neural News Recommendation with Multi-Head Self-Attention.
From chatter to headlines: harnessing the real-time web for personalized news recommendation
References
Machine learning in automated text categorization
What is Twitter, a social network or a news media?
Fab: content-based, collaborative recommendation
Why we twitter: understanding microblogging usage and communities
Content-based recommendation systems
Related Papers (5)
Frequently Asked Questions (9)
Q2. What have the authors stated for future works in "Terms of a feather: content-based news recommendation and discovery using twitter!" ?
There are many opportunities for further work within the scope of this research. Some suggestions include considering preference rankings and click-thrus as part of the recommendation algorithm. Also, it will be interesting to consider whether the reputation of users on Twitter has a bearing on how useful their tweets are during ranking.
Q3. What is the purpose of the paper?
In this paper, the authors consider trending and emerging topics on user-generated content sites like twitter as a way to automatically derive recommendation data for topical news and web-item discovery.
Q4. How many unique stories were generated during the evaluation period?
In addition, the 35 users registered a total of 281 unique RSS feeds as story sources and during the evaluation period these feeds generated a total of 31,137 unique stories/articles.
Q5. What is the role of the user’s RSS feed?
Buzzer itself is developed as a web application and can take the place of a user’s normal RSS reader: the user continues to have access to their favourite RSS feeds but in addition, by syncing Buzzer with their Twitter account, they have the potential to benefit from a more informative ranking of news stories based on their inferred interests.
Q6. What is the meaning of the term content-based recommender?
The authors are interested in the potential to use near-ubiquitous user-generated content as a source of preference and profiling information in order to drive recommendation, as such in this research context Buzzer is termed a content-based recommender.
Q7. How many tweets were retrieved from the social graphs of the 35 registered users?
During this timeframe the authors gathered a total of 56 million public tweets (for use in strategies S1 and S3) and 537,307 tweets from the social graphs of the 35 registered users (for use in S2 and S4).
Q8. How many click-thrus are expected to be generated by the strategy?
the authors can see that drawing stories from the larger community of RSS feeds (S3 + S4) attracts fewer click-thrus (approximately 150) than stories that are drawn from the user’s personal RSS feeds (strategies S1 + S2), which attract about 225 click-thrus, which is acceptable and expected.
Q9. What is the definition of the space of documents?
The space of documents themselves could be defined by Brusilovsky and Henze as an Open Corpus Adaptive Hypermedia System in that there is an open corpus of documents (though topic specific), that can constantly change and expand [3].