| Morning Session Session Chairs - Jon Degenhardt & Owen Phelan | |
|---|---|
| 8:55 am | Opening Remarks |
| 9:00 am |
Keynote Modeling Explicit and Implicit User Behavior for Finding Answers on the Web Eugene Agichtein Emory University, USA Abstract: Billions of people interact with Web search engines and online forums. These data can be used to derive models of searcher intent, attention, and satisfaction, and, in turn to improve question answering tasks, such as candidate passage retrieval, answer selection, and presentation. I will describe how explicit interactions captured using lightweight instrumentation of both search- and landing pages can be converted to attention and satisfaction signals. Then, I will discuss how these signals can be used to improve selection of relevant documents, passages, and answers, and our initial work on adapting these ideas to new interaction modalities. |
| 9:50 am | Contributed Talk Multi-objective Relevance Ranking Michinari Momma, Alireza Bagheri Garakani and Yi Sun |
| 10:10 am | Contributed Talk Learning Embeddings for Product Size Recommendations Kallirroi Dogani, Matteo Tomassetti, Saúl Vargas, Ben Chamberlain and Sofie de Cnudde |
| 10:30 am | Coffee Break |
| 11:00 am |
Contributed Talk Leverage Implicit Feedback for Context-aware Product Search Keping Bi, Choon Hui Teo, Yesh Dattatreya, Vijai Mohan and W. Bruce Croft |
| 11:20 am |
Short Talks: Ideas & Insights Omar Alonso Microsoft, USA Perspectives on search for e-commerce Abstract: Searching for products and services in online websites is one of the most common activities performed on the Web. Compared to traditional web search scenarios, things are somewhat different in e-commerce. In this talk, I’ll show some perspectives on information retrieval and related areas in the context of e-commerce. David Carmel Amazon, Israel On the Relation between Products' Relevance and Customers' Satisfaction in Voice Shopping Abstract: An emerging usage of intelligent voice assistants is shopping. In the traditional usage scenario, customers issue a product search query by voice, and get back products which they can purchase, or upon which they can take some engagement actions such as add-to-cart or send-to-phone. Looking at customers' behavior in voice shopping, we have observed an interesting and surprising phenomenon, where users purchase or engage with irrelevant search results. In this talk, we analyze this phenomenon and demonstrate its significance. We provide several hypotheses as to the reasons behind it, including customers' preferences, trendiness of the products, their relatedness to the query, the tolerance level of the customer, the query intent and the product price. We address each hypothesis by analyzing the data accordingly and provide insights with respect to customers' purchase and engagement behavior with irrelevant results. Owen Phelan Zalando, Ireland Getting people to describe fashion is hard Abstract: In this short talk I’ll give an overview of how Zalando - one of the biggest fashion e-commerce companies in Europe - works with Product Data. In particular we will focus on the challenges we face in describing fashion products, and how these can have impact across the entire company and customer experience. |
| 12:30 pm | Lunch Break |
| Afternoon Session Session Chair - Andrew Trotman | |
| 1:30 pm |
Keynote Building a Broad Knowledge Graph for Products Xin Luna Dong Amazon, USA Abstract: Knowledge graphs have been used to support a wide range of applications and enhance search results for multiple major search engines, such as Google and Bing. At Amazon we are building a Product Graph, an authoritative knowledge graph for all products in the world. The thousands of product verticals we need to model, the vast number of data sources we need to extract knowledge from, the huge volume of new products we need to handle every day, and the various applications in Search, Discovery, Personalization, Voice, that we wish to support, all present big challenges in constructing such a graph. In this talk we describe our efforts in building a broad product graph, a graph that starts shallow with core entities and relationships, and allows easily adding verticals and relationships in a pay-as-you-go fashion. We describe our efforts on knowledge extraction, linkage, and cleaning to significantly improve the coverage and quality of product knowledge. We also present our progress towards our moon-shot goals including harvesting knowledge from the web, hands-off-the-wheel knowledge integration and cleaning, human-in-the-loop knowledge learning, and graph mining and graph-enhanced search. |
| 2:15 pm | Panel Discussion Ecommerce Discovery vs Web Search: Same or Different? Vanessa Murdock Amazon, USA Estelle Afshar The Home Depot, USA David Carmel Amazon, Israel Charles Clarke University of Waterloo, Canada Xin Luna Dong Amazon, USA |
| 3:00 pm | Coffee Break |
| 3:30 pm |
Report on SIGIR eCom'19 High Accuracy Recall Task Jon Degenhardt |
| 3:50 pm |
Group Discussion |
| 4:05 pm |
Poster Session |
| 5:30 pm | Closing |