University of Cambridge > Talks.cam > NLIP Seminar Series > Can we optimise fashion ecommerce using NLG?

Can we optimise fashion ecommerce using NLG?

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact James Thorne.

Note: earlier time of 11am

In this talk we describe the research behind emotif.ai, which is a new NLG startup. Emotif.ai’s technology is derived from an earlier work where we built a metaphor generation system.

The original research goal was the generation of novel and apt metaphors for a given context. This project required that we compile datasets, annotate for metaphors, build models, and perform human evaluation. We discuss the challenges associated with each of these stages and describe a novel heterogeneous encoder-decoder architecture that generates many realisations for the same context.

Emotif.ai’s system uses a similar architecture to generate many descriptions for a single garment. The different descriptions emphasise emotional tones, which allows for A/B testing of copy variants. Finally, we discuss some of the lessons learnt commercialising a research project.

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2020 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity