A heuristic ranking of different characteristic mining based mathematical formulae retrieval models

Abstract

The significant difficulty in the present circumstances is how to classify the math related keywords from a given text file and group them in one math file. Through this article a heuristic ranking model was developed and was evaluated on different mathematical formulae retrieval algorithms based on Characteristic mining. Our proposed heuristic ranking model was developed using the performance metrics of exiting retrieval algorithms such as NMF clustering, Levenstein distance, Sequence matcher, Fuzzy-wuzzy and Tensorflow. Performance metrics such as sensitivity, specificity, efficiency, accuracy and retrieval time were used in developing our heuristic ranking model.