NextSmarty joins as speaker in the top conference on Recommender Systems RecSys 2017 (Como)
About RecSys: the number one conference on recommender systems
The ACM Recommender Systems conference (RecSys) is the premier international forum for the presentation of new research results, systems and techniques in the broad field of recommender systems. Recommendation is a particular form of information filtering, that exploits past behaviors and user similarities to generate a list of information items that are personally tailored to an end-user’s preferences. As RecSys brings together the main international research groups working on recommender systems, along with many of the world’s leading e-commerce companies, it has become the most important annual conference for the presentation and discussion of recommender systems research. RecSys 2017, the eleventh conference in this series, was held in Como, Italy. Participants—more than 600—came from academia and industry presenting their latest results and identify new trends and challenges in providing recommendation components in a range of innovative application contexts. In addition to the main technical track, RecSys 2017 program featured keynotes and invited talks, tutorials covering state-of-the-art in this domain, a workshop program, an industrial track and a doctoral symposium.
NextSmarty’s presentation at RecSys conference
In many real-life recommendation settings, user profiles and past activities are not available. The recommender system should make predictions based on session data, e.g. session clicks and descriptions of clicked items. Conventional recommendation approaches, which rely on user-item interaction data, cannot deliver accurate results in these situations.
In this paper, we describe a method that combines session clicks and content features such as item descriptions and item categories to generate recommendations. To model these data, which are usually of different types and nature, we use 3-dimensional convolutional neural networks with character-level encoding of all input data.
While 3D architectures provide a natural way to capture spatio-temporal patterns, character-level networks allow modeling different data types using their raw textual representation, thus reducing feature engineering effort. We applied the proposed method to predict add-to-cart events in e-commerce websites, which is more difficult than just predicting next clicks. On two real data sets, our method outperformed several baselines and a state-of-the-art method based on recurrent neural networks.
NextSmarty speaks in session 4 https://recsys.acm.org/recsys17/session-4/
3D Convolutional Networks for Session-based Recommendation with Content Features
by Trinh Xuan Tuan and Tu Minh Phuong
Sign up with us and keep updated for the next post – we will share our detail idea about our implementation.
11th ACM Conference on Recommender Systems
Como, Italy, 27th-31st August 2017