Criteo is a "personalized retargeting company" that works with Internet retailers to serve personalized online display advertisements to consumers that have previously visited the advertiser's website. The company currently operates in a total of 30 markets around the world and is headquartered in Paris, France. On April 7, 2011, Criteo announced that it hired Greg Coleman as president. Previously, Coleman served as president and chief revenue officer of The Huffington Post and executive vice president of global sales for the company Yahoo!.Criteo enables online businesses to follow up visitors who have left their website without making a purchase using personalized banners which aim to drive potential customers back to the business website. Wikipedia.
Agarwal A.,Microsoft |
Chapelle O.,Criteo |
Dudik M.,Microsoft |
Journal of Machine Learning Research | Year: 2014
We present a system and a set of techniques for learning linear predictors with convex losses on terascale data sets, with trillions of features,1 billions of training examples and millions of parameters in an hour using a cluster of 1000 machines. Individually none of the component techniques are new, but the careful synthesis required to obtain an efficient implementation is. The result is, up to our knowledge, the most scalable and efficient linear learning system reported in the literature.2 We describe and thoroughly evaluate the components of the system, showing the importance of the various design choices. © 2014 Alekh Agarwal, Olivier Chapelle, Miroslav Dudík and John Langford. Source
Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining | Year: 2014
In performance display advertising a key metric of a campaign effectiveness is its conversion rate - the proportion of users who take a predefined action on the advertiser website, such as a purchase. Predicting this conversion rate is thus essential for estimating the value of an impression and can be achieved via machine learning. One difficulty however is that the conversions can take place long after the impression - up to a month - and this delayed feedback hinders the conversion modeling. We tackle this issue by introducing an additional model that captures the conversion delay. Intuitively, this probabilistic model helps determining whether a user that has not converted should be treated as a negative sample - when the elapsed time is larger than the predicted delay - or should be discarded from the training set - when it is too early to tell. We provide experimental results on real traffic logs that demonstrate the effectiveness of the proposed model. © 2014 ACM. Source
Criteo | Date: 2015-10-05
Systems and methods for generating and enhancing advertising campaigns and web content using dominant attribute analysis in accordance with embodiments of the invention are disclosed. In one embodiments, a product advertising server system includes a processor and a memory connected to the processor and storing a product advertisement generation application, wherein the product enhancement generation application directs the processor to obtain product feed data, determine category data based on the obtained product feed data, identify dominant attribute data based on the category data, generate a set of product title data based on the dominate attribute data and the product feed data, and store the generated product title data using the memory.
Criteo | Date: 2013-07-15
Described are methods, systems, and apparatus, including computer program products for domain selection for advertisement data in the delivery of website display ads. A request is received from a requester for indicia of one or more impression opportunities. A redirection URL and redirection type is received. One or more desired impression opportunities are determined. The indicia of the one or more impression opportunities are generated comprising indicia of the one or more desired impression opportunities. Redirection instructions are generated based on the redirection URL and redirection type. The indicia of the one or more impression opportunities and the redirection instructions are sent to the requester.
Criteo | Date: 2013-08-13
A first request on a second domain associated with the computing device is received, by the computing device, from a user device, in response to the user device processing a webpage associated with a first domain. Handler instructions are sent, by the computing device, to the user device. A second request including a target URL is received, by the computing device, from the user device. Setting instructions to set a first cookie on the second domain associated with the computing device and redirection instructions to redirect the user device to the target URL are sent, by the computing device, to the user device.