Presentation on theme: "Learning to Question: Leveraging User Preferences for Shopping Advice Date : 2013/12/11 Author : Mahashweta Das, Aristides Gionis, Gianmarco De Francisci."— Presentation transcript:
Learning to Question: Leveraging User Preferences for Shopping Advice Date : 2013/12/11 Author : Mahashweta Das, Aristides Gionis, Gianmarco De Francisci Morales, and Ingmar Weber Source : KDD13 Advisor : Jia-ling Koh Speaker : Yi-hsuan Yeh
Introduction 3 Motivation Customers shop online, from their homes, without any human interaction involved. Catalogs of online shops are so big and with so many continuous updates that no human, however expert, can effectively comprehend the space of available products. Use a flowchart asks the shopper a question, and the sequence of answers leads the shopper to the suggested shopping option.
Introduction 4 S HOPPING A DVISOR is a novel recommender system that helps users in shopping for technical products. car
Introduction 5 S HOPPING A DVISOR generates a tree-shaped flowchart, in which the internal nodes of the tree contain questions involve only attributes from the user space. non-expert users can understand easily.
Introduction 6 1. How to learn the structure of the tree, i.e., which questions to ask at each node. Find the best user attribute to ask at each node. * This paper focus on identifying the attribute of interest, and not on the task of formulating the question in a human interpretable way. 2. How to produce a suitable ranking at each node. Learning-to-rank approach
Outline Introduction Method – L EARN SAT REE algorithm Experiments Conclusion 7
L EARN SAT REE algorithm 8 1. Table U (user) 2. Table P (product) 3. Table R (review) attributes users
*User attributes 9 1. Car (from Yahoo! Autos) Ex fuel economy, comfortable interior, stylish exterior 2. Camera (form Flickr) Photos tag topic Ex food topic (tags fruit, market)
Problem definition Build tree 2. Rank products Top-k list of product recommendations
Learning product rankings 11 A > B B > C B > D. R ANK SVM model R ANK SVM model ABDC...ABDC... features Products technical attributes
12 a1a2a3a4a5 Product A10111 Product B10010
Learning the tree structure 13
14 System result
15 user split
16 product R ANK SVM Count payoff ABDC...ABDC... FBEA...FBEA... Rank list
Stopping criterion 17 1) Grow the tree to its entirety 2) Post-pruning If a nodes child node is split by the near-synonomous tag trim the child node Example: travel vacation Employ pruning rules on the validation set.
Experiment setup S HOPPING A DVISOR Authors method 2. R ANK SVM The ranked list returned by S HOPPING A DVISOR at the root 3. k-NN k-nearest neighbors algorithm 4. SA.k-NN Features are selected from S HOPPING A DVISOR
Quality evaluation 21 average MRR ABD...ABD... System result ranking list 25 topics12 topics
Conclusion Proposed a novel recommender system, S HOPPING A DVISOR, that helps users to shop for technical products. S HOPPING A DVISOR leverages both user preferences and technical product attributes in order to generate its suggestions. At each node, S HOPPING A DVISOR suggests a ranking of products matching the preferences of the user. Compared with a baseline, and demonstrated the effectiveness of the approach. 24