When shopping online, buyers often express and refine their purchase preferences by exploring different items in the product catalog based on varying attributes, such as color, size, shape, and material. As such, it is increasingly important for e-commerce ranking systems to quickly learn a buyer’s fine-grained preferences and re- rank items based on their most recent activity within the session. In this paper, we propose an 𝑂nline 𝑃ersonalized 𝐴ttribute-based 𝑅e-ranker (OPAR), a light-weight, within-session personalization approach using multi-arm bandits (MAB). By representing each arm in the MAB as an attribute, we reduce the complexity space while offering more fine-grained personalization.

E-commerce platforms offer the convenience of browsing through an entire catalog of inventory via a search bar. An unconventional inventory of unique products presents even greater challenges for product search, given that many of listings fall outside of standard e-commerce categories. With the potentially overwhelming number of relevant items per query, it becomes increasingly important for market places and platforms to help the user find items that best fit their preference and interest via a user-centric ranking model that generates personalized search results.

To discover such latent intents, and use them effectively for recommendation, in this paper we propose an Attentive Sequential model of Latent Intent (ASLI in short). Our model first learns item similarities from users' interaction histories via a self-attention layer, then uses a Temporal Convolutional Network layer to obtain a latent representation of the user's intent from her actions on a particular category. We use this representation to guide an attentive model to predict the next item.