Learning Latent Vector Spaces for Product Search
This work addresses product search for e-commerce users, but it is incremental as it builds on existing latent vector space methods with a novel joint learning approach.
The authors tackled product search in e-commerce by introducing a latent vector space model that jointly learns representations of words and products without explicit annotations, achieving enhanced performance in learning-to-rank evaluations compared to existing models like LSI, LDA, and word2vec.
We introduce a novel latent vector space model that jointly learns the latent representations of words, e-commerce products and a mapping between the two without the need for explicit annotations. The power of the model lies in its ability to directly model the discriminative relation between products and a particular word. We compare our method to existing latent vector space models (LSI, LDA and word2vec) and evaluate it as a feature in a learning to rank setting. Our latent vector space model achieves its enhanced performance as it learns better product representations. Furthermore, the mapping from words to products and the representations of words benefit directly from the errors propagated back from the product representations during parameter estimation. We provide an in-depth analysis of the performance of our model and analyze the structure of the learned representations.