Creation of an Accurate Model of the Topical Structure of PubMed and associated indicators

NIH through NIDA seeks development of an advanced and sophisticated analytical tool and approaches to enhance the professional evaluation and decision support in life sciences management and administrationIn this Phase II project the Contractor will continue to develop indicators of impactful science and of how virtuous the research ise gtransparencyreproducibility and potential for translation applicationand correlate them with existing citation based indicatorsIn previous studiesthe Contractor created a methodology to partition records from PubMed into scientific topics that are just as accurate as those that can be identified from citation dataThe ability to identify emerging topics is one of the most important tasks in developing a research strategyThe ability to determine if a topic is scientifically attractive and potential for future research is of comparable importance for portfolio analysisOverallthe accuracy by which one identifies topicsand the validity of any indicator of scientific topic attractivenessis a necessary first step towards research evaluation and planning that is evidence basedThis need is not adequately addressed in a cost efficient manner by any existing productThe current market for these types of indicators is being served by the three companies with citation databasesClarivateWeb of ScienceElsevierScopusand Springer NatureDimensionsThe ultimate objective of this project is to develop an accurate and viable open source alternative to citation based indicators of science and technology that can be used for research assessment and funding decisions