This paper addresses the problem of information burying in social sciences, where a large amount of experimental findings reported in multiple scientific articles may be missed by scholars due to the lack of an active accumulation, organization and synthesis of these findings into a centralized information system. To tackle the information burying problem, in this paper we present a new network-based data model and methodology for aggregating, organizing, linking and mining quantitative results published in multiple academic articles in particular sub-fields of social sciences. The goal of the proposed methodology is to provide researchers with a wider perspective when viewing scientific results in their own fields and utilize it for their research. To validate the proposed approach, we conducted a manual experiment with a corpus of 41 scientific articles in the field of personal information management. The experiment indicates that the constructed network-based information system can be effectively used to explore the relationships between the results of various articles, raising new research questions and hypotheses based on results from multiple articles that tested similar variables. The proposed system can serve as a catalyst for the advancement of research in various fields of social science.
Bibliographical notePublisher Copyright:
© 2020, Akadémiai Kiadó, Budapest, Hungary.
- Buried information
- Data mining
- Network of variables
- Social science
- Statistical relationships