Using hybrid algorithmic-crowdsourcing methods for academic knowledge acquisition

摘要: Scientific literature contains a lot of meaningful objects such as Figures, Tables, Definitions, Algorithms, etc., which are called Knowledge Cells hereafter. An advanced academic search engine could take advantage and their various relationships to obtain more accurate results is expected. Further, it’s expected provide fine-grained regarding for deep-level information discovery exploration. Therefore, it important identify extract the often intrinsic implicit in articles. With exponential growth scientific publications, acquisition useful knowledge impose some practical challenges For example, existing algorithmic methods can hardly extend handle diverse layouts journals, nor scale up process massive documents. As crowdsourcing has become powerful paradigm large problem-solving especially tasks that difficult computers but easy human, we consider problem crowd-sourced database show hybrid framework integrate accuracy workers speed automatic algorithms. In this paper, introduce our current system implementation, platform (PANDA), well interesting observations promising future directions.