similarity

Predicting human similarity judgments with distributional models: The value of word associations

When do learned transformations influence similarity and categorization?

Predicting human similarity judgments with distributional models: The value of word associations

Finds that semantic networks built from word association data capture similarity judgments better than state-of-the-art corpus models like word2vec.

Structure at every scale: A semantic network account of the similarities between very unrelated concepts

Evidence for widespread thematic structure in the mental lexicon

Strong structure in weak semantic similarity: A graph based account