The goals of the project are:
- to develop new text mining unsupervised methods for deriving term similarities, based on distributional semantics, to produce meaningful and high quality document and label clusters to support screen while searching in EBPH reviews.
- to develop new seariation algorithms for ranking and visualising meaningful associations of multiple types, dynamically and iteratively.
- to evaluate these newly developed methods in EBPH reviews, based on implementation of a pilot, to ascertain the level of transformation in EBPH reviewing.
One of the outcomes of the project is RobotAnalyst : an online system to support searching and screening in systematic reviews.