Unsupervised Does Not Mean Uninterpretable: The Case for Word Sense Induction and Disambiguation

By Chris Biemann, Simone Paolo Ponzetto, Stefano Faralli, Alexander Panchenko, Eugen Ruppert. 2017

35 citations recorded by the source corpus.

The current trend in NLP is the use of highly opaque models, e.g. neural networks and word embeddings. While these models yield state-of-the-art results on a range of tasks, their drawback is poor interpretability. On the example of word sense induction and disambiguation (WSID), we show that it is possible to develop an interpretable model that matches the state-of-the-art models in accuracy. Namely, we present an unsupervised, knowledge-free WSID approach, which is interpretable at three levels: word sense inventory, sense feature representations, and disambiguation procedure. Experiments show that our model performs on par with state-of-the-art word sense embeddings and other unsupervised systems while offering the possibility to justify its decisions in human-readable form.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

What do we need to build explainable AI systems for the medical domain? (2017, background): “Combined with an interpretable disambiguation system (Panchenko et al., 2017), this realizes concept-linking in context with high accuracy while providing human-interpretable reasons for why concepts have been selected.”
From Word to Sense Embeddings: A Survey on Vector Representations of Meaning (2018, background): “…Pilehvar, Camacho-Collados, Navigli, & Collier, 2017), document similarity (Wu & Giles, 2015), and word sense induction (Pelevina et al., 2016; Panchenko, Ruppert, Faralli, Ponzetto, & Biemann, 2017b) and disambiguation (Chen et al., 2014; Rothe & Sch¨utze, 2015; Camacho-Collados et al.,…”
Explainable Artificial Intelligence: a Systematic Review (2020, background): “Eventually, Unsupervised Interpretable Word Sense Disambiguation [318] produces interpretable word sense disambiguation models that create clusters, or inventories, of words.”
Embedding Words and Senses Together via Joint Knowledge-Enhanced Training (2016, background): “Additionally, their induced senses are neither readily interpretable (Panchenko et al., 2017) nor easily mappable to lexical resources, which limits their application.”
Explainable Sentiment Analysis with Applications in Medicine (2018, background): “model is built continuously and improves over time via feedback and interaction of domain experts and, ii) an interpretable disambiguation system that can provide common sense and human-readable reasons why in a given context, a given sense was detected [79].”