Unsupervised Word Sense Disambiguation Using Markov Random Field and Dependency Parser

Authors: Devendra Chaplot, Pushpak Bhattacharyya, Ashwin Paranjape

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our graph-based unsupervised WSD system beats state-of-the-art system on Sens Eval-2, Sens Eval-3 and Sem Eval-2007 English all-words datasets while being over 35 times faster.
Researcher Affiliation Collaboration Devendra Singh Chaplot Samsung Electronics Co., Ltd. Suwon, South Korea Pushpak Bhattacharyya Computer Science Department IIT Bombay, India Ashwin Paranjape Stanford University California, US
Pseudocode No The paper includes a block diagram (Figure 1) but no structured pseudocode or algorithm blocks.
Open Source Code No Section 8 states that the web interface "shall be made open-source", indicating future availability, not current. No concrete link is provided for the methodology's code.
Open Datasets Yes We have tested our system on the Sens Eval-2 (Palmer et al. 2001), Sens Eval-3 (Snyder and Palmer 2004) and Sem Eval-2007 (Pradhan et al. 2007) English all-words WSD datasets.
Dataset Splits No The paper does not provide specific details about train, validation, or test splits. It states that it was tested on the full datasets.
Hardware Specification Yes MRF-LP 2.4Ghz 8GB 1144 MRF-SP 2.4Ghz 8GB 2698
Software Dependencies No The paper mentions using "Matlab UGM (Schmidt 2007) package", "Princeton Word Net", "Link Parser", "Stanford Parser", and "Stanford POS Tagger" but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes We have used Path relatedness measure for our experiments.