2017/05/12 by Dawn Chen, Joshua C. Peterson, Chen, Dawn +3 · 31 citations
Computer Science · Mathematics · Neuroscience · #Analogy #Biology #Cognitive Science and Education Research #Computability, Logic, AI Algorithms #Computation and Language (cs.CL) #Computer science #Epistemology #FOS: Computer and information sciences #Mathematics #Neural Networks and Applications #Philosophy #Space (punctuation) #Vector (molecular biology) #cs.CL
paper · pdf · doi:10.48550/arxiv.1705.04416
published in arXiv (Cornell University) (Cornell University) · 6 pages, 4 figures, In the Proceedings of the 39th Annual Conference of the Cognitive Science Society
openalex publication_date 2017/05/12 · arxiv created 2017/06/08 · arxiv updated 2017/06/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Vector-space representations provide geometric tools for reasoning about the similarity of a set of objects and their relationships. Recent machine learning methods for deriving vector-space embeddings of words (e.g., word2vec) have achieved considerable success in natural language processing. These vector spaces have also been shown to exhibit a surprising capacity to capture verbal analogies, with similar results for natural images, giving new life to a classic model of analogies as parallelograms that was first proposed by cognitive scientists. We evaluate the parallelogram model of analogy as applied to modern word embeddings, providing a detailed analysis of the extent to which this approach captures human relational similarity judgments in a large benchmark dataset. We find that that some semantic relationships are better captured than others. We then provide evidence for deeper limitations of the parallelogram model based on the intrinsic geometric constraints of vector spaces, paralleling classic results for first-order similarity.