CLJul 12, 2017
A Critique of a Critique of Word Similarity Datasets: Sanity Check or Unnecessary Confusion?
arXiv:1707.03819v11 citations
Originality Synthesis-oriented
AI Analysis
This is an incremental critique for researchers in computational lexical semantics.
The paper critiques a previously proposed sanity check for evaluating word similarity datasets, arguing that the test is unstable and offers no added insight.
Critical evaluation of word similarity datasets is very important for computational lexical semantics. This short report concerns the sanity check proposed in Batchkarov et al. (2016) to evaluate several popular datasets such as MC, RG and MEN -- the first two reportedly failed. I argue that this test is unstable, offers no added insight, and needs major revision in order to fulfill its purported goal.