Expertise-aware news feed updates recommendation: a random forest approach

摘要

With social media being widely used around the world, and because of the large amount of data, users are overcome by updates displayed chronologically in their news feed. Furthermore, most updates are considered irrelevant. To help beneficiary users quickly catch up with the relevant content, ranking news feed updates in descending relevance order has been achieved based on the prediction of a relevance score between a beneficiary and a new update in the news feed. Four types of features are generally used to predict the relevance: (1) the relevance of the update content to the beneficiary's interests; (2) the social tie strength between the beneficiary and the update's author; (3) the author's authority; and (4) the update quality. In this work, from the biography and the textual content posted, we propose an approach that infers and uses another type of feature which is the expertise of the update's author for the corresponding topic. Following extensive experiments on a real dataset crawled from Twitter, the results show that infer the author's expertise is critical for identifying relevant updates in news feeds.