Robustness Measurement of Features using Various Machine Learning Methods

Acronyms and expansions are often used in written media such as articles, abstracts, newspapers, and documents. However, making a computer capable of automatically recognizing pairs of acronyms and expansions is challenging. In this study, we evaluate the reliability of the eight features introduced in our previous study by employing various supervised learning methods such as Naive Bayes, Random Forest, AdaBoost, and gradient boosting. The results show that the Random Forest algorithm achieves the highest accuracy (97.65%) with 97.72% precision, 97.65% recall, and 97.65% F-measure.