implementation of algorithms which compute sentence similarities
similarity_ideas.py Use the functions in similarity_ideas.py to compute similarity matrices between texts.
data_reader.py Contains functions that read data into a format processible by similarity_ideas.py
preselect_ideas_for_manual_annotation.py Implements the function of the sem_eval paper 2017 to preselect ideas "worthy" of being annotated by crowd workers. All other texts are assumed to have a similarity of zero.
sts_evaluation.py Calculates the similarity between manual annotations and a similarity file. Currently only for sem_eval 2017.
experiment_....py Files that start with experiment_... are for experiments.