How to use the thousand words of an image to power a fashion shopping chatbot
As the chatbot space grows and matures, more and more fashion retailers are using chatbot platforms to drive shopping. In mode.ai, we developed a general infrastructure to power a rich kind of shopping chatbots, where the user can interact through text, images and buttons. Understanding the user's intent and search content are crucial for providing a smooth user experience. In my talk, I will explain how deep-learning analysis of images uploaded by the user is combined with NLP analysis of user's text input, prior knowledge about the user and other inputs to determine the user's intent.
Or Cohen is the Head of Engineering and AI at mode.ai, a startup that powers fashion shopping through chat platforms for retailers and consumers. Together with a small team of engineers, Or is responsible for the development of the engineering infrastructure and algorithms that power mode.ai's visual chatbot. Or has a Ph.D. in Physics from the Weizmann Institute of Science. His studies were focused on statistical physics and numerical methods.