We present two systems created for SemEval2016s Task 11: Complex Word Identification. Our two systems, a regression tree and decision tree, were trained with a word’s unigram and lemma word counts, average ageof-acquisition, and a measure of concreteness. The systems ranked 5th and 6th, respectively, on the test set by G-score (the harmonic mean between accuracy and recall). With the regression tree’s predictions earning a G-score of 0.766, and the decision tree’s earning 0.765, the two systems scored within 1 percent of the score of the best-performing system in the task.
Paper
Full text
HMC at SemEval-2016 Task 11: Identifying Complex Words Using Depth-limited Decision Trees
Semantic Scholar · Computer Science · 2016
Abstract
We present two systems created for SemEval2016s Task 11: Complex Word Identification. Our two systems, a regression tree and decision tree, were trained with a word’s unigram and lemma word counts, average ageof-acquisition, and a measure of concreteness. The systems ranked 5th and 6th, respectively, on the test set by G-score (the harmonic mean between accuracy and recall). With the regression tree’s predictions earning a G-score of 0.766, and the decision tree’s earning 0.765, the two systems scored within 1 percent of the score of the best-performing system in the task.