Feature-Based Researcher Identification Framework Using Timeline Data

Studies are actively ongoing for better understanding and strengthening the capabilities of researchers. To do so requires an accurate diagnosis and analysis of such researchers. Therefore, data of each researcher must be collected and be identified in a big-data environment. Consequently, researcher-name identification has emerged as an important issue. This paper proposes a framework for collecting, refining, identifying, and publicly offering researcher data. For identifying authors’ name, the proposed framework extracts timeline based patterns that make help to identify the same name authors with their representative attributes such as emails and affiliations. The results of the proposed framework based on timeline patterns, show a 69.5 % average author-identification rate given a group of otherwise unidentified authors.