We study the problem of automatically building hypernym taxonomies from textual and visual data. Previous works in taxonomy induction generally ignore the increasingly prominent visual data, which encode important perceptual semantics. Instead, we propose a probabilistic model for taxonomy induction by jointly leveraging text and images. To avoid hand-crafted feature engineering, we design end-to-end features based on distributed representations of images and words. The model is discriminatively trained given a small set of existing ontologies and is capable of building full taxonomies from scratch for a collection of unseen conceptual label items with associated images. We evaluate our model and features on the WordNet hierarchies, where our system outperforms previous approaches by a large gap.
@article{arxiv.1606.09239,
title = {Learning Concept Taxonomies from Multi-modal Data},
author = {Hao Zhang and Zhiting Hu and Yuntian Deng and Mrinmaya Sachan and Zhicheng Yan and Eric P. Xing},
journal= {arXiv preprint arXiv:1606.09239},
year = {2016}
}