Abstract
In this paper, the authors present a method to deriving type of scientific results from the texts of scientific articles and implementing these results into recommender systems to enhance their output. Type of scientific results and text data are divided into corresponding classes, which are based on hypothetical user needs. The classes indicate the type of a scientific result in article. The proposed approach involves the determination of meaningful collocations for scientific article. Topic modeling is applied to the received collocation sets from the text of an article abstract. The topics (classes) obtained show the relationship between a scientific article and the type of scientific result.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
