Search

As online video service continues to grow in popularity, video content providers compete hard for more eyeball engagement. Some users visit multiple video sites to enjoy videos of their interest while some visit exclusively one site. However, due to the isolation of data, mining and exploiting user behaviors in multiple video websites remain unexplored so far. In this work, we try to model user preferences in six popular video websites with user viewing records obtained from a large ISP in China. The empirical study shows that users exhibit both consistent cross-site interests as well as site-specific interests. To represent this dichotomous pattern of user preferences, we propose a generative model of Multi-site Probabilistic Factorization ( MPF ) to capture both the cross-site as well as site-specific preferences. Besides, we discuss the design principle of our model by analyzing the sources of the observed site-specific user preferences, namely, site peculiarity and data sparsity. Through conducting extensive recommendation validation, we show that our MPF model achieves the best results compared to several other state-of-the-art factorization models with significant improvements of F-measure by 12.96, 8.24 and 6.88 percent, respectively. Our findings provide insights on the value of integrating user data from multiple sites, which stimulates collaboration between video service providers.

The academic community has seen a tremendous growth in number of authors and publications in recent times. Most previous studies of the academic community, whether on individual activity, productivity and influence, or collaboration patterns and their implications, have been based on a flat and static view of the system. However, the academic community resembles a dynamic growing population, with entry and exit of authors. In this paper, we study this systemic inflation by proposing a population model for academia. We use a generalized branching process as an overarching framework, which enables us to describe the research community from an evolutionary and structural perspective. Further, the observed patterns allow us to shed light on researchers’ life-cycle encompassing their arrival, academic life expectancy , activity, productivity and offspring distribution in the community. In our study, we used data from DBLP for a case study in the computer science community, although our methodology can be adopted in a systematic manner to any other research domain with sufficient publication records. We believe that the results can help academics and policy makers alike to better understand and evaluate the development and evolution of their respective academic communities.