Abstract
A detailed understanding of users contributes to the understanding of the Web's evolution, and to the development of Web applications. Although for new Web platforms such a study is especially important, it is often jeopardized by the lack of knowledge about novel phenomena due to the sparsity of data. Akin to human transfer of experiences from one domain to the next, transfer learning as a subfield of machine learning adapts knowledge acquired in one domain to a new domain. We systematically investigate how the concept of transfer learning may be applied to the study of users on newly created (emerging) Web platforms, and propose our transfer learning-based approach, TraNet. We show two use cases where TraNet is applied to tasks involving the identification of user trust and roles on different Web platforms. We compare the performance of TraNet with other approaches and find that our approach can best transfer knowledge on users across platforms in the given tasks.
Highlights
The Web evolves in a permanent cycle (Fig. 1), as portrayed by Hendler et al [1]: An idea may lead to novel technology as well as social activities
Since structural features are by nature present in all networks, in order to better investigate the principle idea behind transfer learning on the Web, we focus on the aspect of structural features
We can achieve the best performance with transfer learning using our approach TraNet
Summary
The Web evolves in a permanent cycle (Fig. 1), as portrayed by Hendler et al [1]: An idea may lead to novel technology as well as social activities. They observed and classified multiple online communities They proposed several measures useful to quantify differences between communities, such as the number of initiative-takers or the length of discussions. The evolution of a Web platform would be greatly facilitated, and the learning cycle would be cut short, if measurements of social behavior could be transferred from previous experiences to new ones, not just based on qualitative observations, and based on quantifiable rules. A new Web platform might want to discourage trolls and encourage trusted users without running through the learning cycle multiple times by transferring quantitative experiences from previous Web platforms. Better than ignoring the heterogeneity, human experts are able to learn from few examples they observe from existing platforms, and transfer their “experience” to new situations. The necessary code (github.com/yfiua/TraNet) and public datasets to reproduce the results in the paper and apply TraNet to accomplish similar tasks are made available online
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
