Combining domain-specific heuristics for author name disambiguation

摘要: Author name disambiguation has been one of the hardest problems faced by digital libraries since their early days. Historically, supervised solutions have empirically outperformed those based on heuristics, but with burden having to rely manually labelled training sets for learning process. Moreover, most just apply some type generic machine solution and do not exploit specific knowledge about problem. In this paper, we follow a similar reasoning, in opposite direction. Instead extending an existing solution, propose set carefully designed heuristics similarity functions supervision only optimize such parameters each particular dataset. As our experiments show, result is very effective, efficient practical author method that can be used many different scenarios.