Multi-Modal Word Synset Induction
Authors: Jesse Thomason, Raymond J. Mooney
IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Both automated and human evaluations demonstrate that multi-modal synset induction outperforms uni-modal induction. More importantly, human judges do not significantly favor Image Net synsets over multi-modal, induced synsets; however, humans do favor Image Net s over uni-modally induced synsets. Automated Evaluation. We computed the v-measure [Rosenberg and Hirschberg, 2007] of the induced synsets, calculated as the harmonic mean of their homogeneity and completeness with respect to the gold-standard Image Net synsets. |
| Researcher Affiliation | Academia | Jesse Thomason and Raymond J. Mooney Department of Computer Science, University of Texas at Austin Austin, TX 78712, USA {jesse, mooney}@cs.utexas.edu |
| Pseudocode | No | The paper describes the methods in narrative text and uses figures for illustration, but it does not include any formal pseudocode blocks or algorithm listings. |
| Open Source Code | Yes | We make this corpus of Image Net synsets associated with text, VGG features, and LSA embedding features per synset observation URL available.3 https://github.com/thomason-jesse/synpol |
| Open Datasets | Yes | We make this corpus of Image Net synsets associated with text, VGG features, and LSA embedding features per synset observation URL available.3 https://github.com/thomason-jesse/synpol |
| Dataset Splits | Yes | We hold out the synsets used to train VGG as validation data in our work. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as GPU/CPU models or memory. |
| Software Dependencies | No | The paper mentions software components like "VGG network" and techniques like "LSA" and "k-means clustering" but does not provide specific version numbers for any software or libraries used. |
| Experiment Setup | Yes | k-means clustering powers polysemy detection, where k is estimated for each set of observations Onp using the gap statistic [Tibshirani et al., 2001]. Intuitively, the gap statistic selects the smallest number of clusters k that reduces within-dispersion compared to k 1 by more than chance. Additionally, we enforce a constraint that no induced sense has fewer than 20 observations (estimated as the mean senses per noun phrase minus one standard deviation in the development data). Greedy merges of the nearest means produces a final set of K induced synsets, R, each of which comprises no more than L distinct word senses. Membership in each induced synset r R is the union of observations of the senses ga . . . gb G whose observations were merged (i.e. r = {ga . . . gb}). K is set based on the ratio of senses to synsets in the development data V (so K fluctuates depending on the number of senses to be clustered). The maximum number of senses per synset, L = 32, is also estimated from V. |