Leverage a standalone text mining software application to prioritize documents for expert review.
Using advanced algorithms from the domains of natural language processing and machine learning, DoCTER greatly improves efficiency in tasks that involve large volumes of text.
Clustering with seeds takes the guesswork out of prioritizing documents—such as abstracts from scientific literature—and generates unbiased forecasts of retrieval accuracy.
DoCTER assigns a reference to a single cluster, providing a topic signature (i.e., set of keywords) for analyzing large bodies of text. Subject matter experts review keywords and assign priority levels to every cluster.
DoCTER is designed to optimize for precision, recall, or F-1 score based on user selections.