LogicAI develops AI for improved hotel recommendationsThe competition was designed to have teams focus on developing algorithms that can improve the quality of hotel recommendations on online platforms. The 500 participating teams analyzed explicit and implicit user signals such as interactions with content on the website or the usage of filter options. The solution, developed by LogicAI, combined different models to better detect the intent of the users which can help influence factors that platforms like trivago can use to increase the relevance of their recommendations.Jens Adamczak, Data Scientist at trivago, on LogicAI’s algorithm: “In the highly competitive market for online travel, fast and relevant recommendation systems are crucial. We are excited about LogicAI’s input and expect to experiment with including parts of the solution into the production system that fuels the offers we present to our users.”Pawel Jankiewicz of LogicAI wins the RecSys Challenge 2019: “From the beginning of the competition, we thought the challenge was very interesting. The problem was well defined and the dataset itself was prepared in an exemplary manner. Although we used traditional feature engineering and Tree Boosting, we are amazed that so many of them used ambitious Deep Learning architectures. It will surely move forward the knowledge of how to built session-based recommendation systems.”Last year’s winner, Layer 6 AI from Toronto, finished second in this years challenge. The teams Meituan-Dianping Group from Beijing, Rosetta AI from Taiwan and letoh govatri from TU Dortmund took third to fifth place.The RecSys Challenge 2019 was organized by trivago, TU Vienna, Polytechnic University Bari and Karlsruhe Institute of Technology. The winners were presented at the 13th ACM RecSys Conference in Copenhagen. The renowned conference on Artificial Intelligence brings together annually the most important international research groups working on recommendation systems as well as many of the world’s leading e-commerce companies.Media contact:
[email protected]This press release contains certain forward-looking statements. Words, and variations of words such as “believe,” “expect,” “plan,” “continue,” “will,” “should,” and similar expressions are intended to identify our forward-looking statements. These forward-looking statements involve risks and uncertainties, many of which are beyond our control, and important factors that could cause actual events and results to differ materially from those in the forward-looking statements. For additional information factors that could affect our forward-looking statements, see our risk factors, as they may be amended from time to time, set forth in our public filings with the Securities and Exchange Commission. We disclaim and do not undertake any obligation to update or revise any forward-looking statement in this press release, except as required by applicable law or regulation.As used herein, references to “we”, “us”, the “company”, or “trivago”, or similar terms shall mean trivago N.V. and, as the context requires, its subsidiariesjens
Bay Street News