Timeline Generation: Tracking individuals on Twitter
This work addresses the challenge of tracking individuals' life events from social media data, which is incremental as it applies existing methods to a new task.
The paper tackles the problem of reconstructing a person's life history by generating a chronological list of personal important events from their tweets, using an unsupervised framework that classifies tweets into four topic types and achieves effectiveness demonstrated on real Twitter data with a new dataset of 20 ordinary users and 20 celebrities.
In this paper, we propose a unsupervised framework to reconstruct a person's life history by creating a chronological list for {\it personal important events} (PIE) of individuals based on the tweets they published. By analyzing individual tweet collections, we find that what are suitable for inclusion in the personal timeline should be tweets talking about personal (as opposed to public) and time-specific (as opposed to time-general) topics. To further extract these types of topics, we introduce a non-parametric multi-level Dirichlet Process model to recognize four types of tweets: personal time-specific (PersonTS), personal time-general (PersonTG), public time-specific (PublicTS) and public time-general (PublicTG) topics, which, in turn, are used for further personal event extraction and timeline generation. To the best of our knowledge, this is the first work focused on the generation of timeline for individuals from twitter data. For evaluation, we have built a new golden standard Timelines based on Twitter and Wikipedia that contain PIE related events from 20 {\it ordinary twitter users} and 20 {\it celebrities}. Experiments on real Twitter data quantitatively demonstrate the effectiveness of our approach.