2014/05/17 by Sunny Mitra, Mitra, Sunny, Ritwik Mitra +9 · 1 citation
Arts and Humanities · Computer Science · Social Sciences · #68T50 #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Language and cultural evolution #Lexicography and Language Studies #Natural Language Processing Techniques #Topic Modeling
paper · pdf · doi:10.48550/arxiv.1405.4392
openalex publication_date 2014/05/17 · openalex created_date 2025/10/24 · openalex updated_date 2026/07/28
In this paper, we propose an unsupervised method to identify noun sense\nchanges based on rigorous analysis of time-varying text data available in the\nform of millions of digitized books. We construct distributional thesauri based\nnetworks from data at different time points and cluster each of them separately\nto obtain word-centric sense clusters corresponding to the different time\npoints. Subsequently, we compare these sense clusters of two different time\npoints to find if (i) there is birth of a new sense or (ii) if an older sense\nhas got split into more than one sense or (iii) if a newer sense has been\nformed from the joining of older senses or (iv) if a particular sense has died.\nWe conduct a thorough evaluation of the proposed methodology both manually as\nwell as through comparison with WordNet. Manual evaluation indicates that the\nalgorithm could correctly identify 60.4% birth cases from a set of 48 randomly\npicked samples and 57% split/join cases from a set of 21 randomly picked\nsamples. Remarkably, in 44% cases the birth of a novel sense is attested by\nWordNet, while in 46% cases and 43% cases split and join are respectively\nconfirmed by WordNet. Our approach can be applied for lexicography, as well as\nfor applications like word sense disambiguation or semantic search.\n