Forecasting technology migrations by means of the technology-topic framework

Technologies such as algorithms, applications and formats usually originate in the context of a specific research area and then spread to several other fields, sometimes with transformative effects. However, this can be a slow and inefficient process, since it not easy for researchers to be aware of all interesting approaches produced by unfamiliar research communities. We address this issue by introducing the Technology-Topic Framework, a novel approach which uses a semantically enhanced technology-topic model and machine learning to forecast the propagation of technologies across research areas. The aim is to foster the knowledge flow by suggesting to scholars technologies that may become relevant to their research field. The system was evaluated on a manually curated set of 1,118 technologies in Semantic Web and Artificial Intelligence and the results of the evaluation confirmed the validity of our approach.