This post is highly inspired from the introduction of a research paper, written by Sébastien Hémon, Thomas Largillier and myself, entitled Partial ranking of products for recommendation systems. Recommending products to customers is a field far older than computer science.  With the tremendous increase of the commercial potential of the e-commerce it has become of the utmost importance to perform well in this field.  Moreover, companies have now the ability to store information about consumption habits of customer: their previous purchases, their tastes. They also have access to information that often allows to cluster customers into communities that share some […]

The category Nice infographics is here to present nice infographics (no kidding!) without comments. I found this one one a french blog, it presents the growth of Internet (click on the image to enlarge it).

Jose Quesada is a cognitive science researcher working at the Adaptive Behavior and Cognition group, Max Planck Institute, Berlin. He’s  writing about productivity and research in a blog entitled « academic productivity« . I cannot really say that I agree with what he writes in his blog. Even if he focuses on productivity in an academic environment, there is no solid justification of why being productive is important. Nevertheless Jose says on his home page:  » I express what I think it is to do conscious science in the 21st century. It may not be what most people think « . I do […]

This is the last of the 3 posts about SpotRank. In this one I show some evidence of the effiency of the method. Even with a very strong analysis of the log files, it is impossible to  judge the quality of the filtering of our method. Indeed, the algorithm consists in filtering news w.r.t. the way people vote, it is not content related. To cope with this issue we decided to gather some feedback from the users themselves.  Since an absolute judgement is impossible to obtain without a long debate on what is the quality of a website, we choose […]

In this post I give a short overview of SpotRank, an algorithm designed by Thomas Largillier, Guillaume Peyronnet and Myself. The goal of SpotRank is to offer a voting mechanism for social news website (such as Digg for instance) which is robust to manipulation attempts by malicious users. Again the post is highly inspired by our research paper. A previous post introduced social news websites and related issues. We consider that the voting system SpotRank is used by a community of users that can propose its own news (or content), that we will call spots. Any user of the community […]

This post is a part of the introduction of an academic paper coauthored by Thomas Largillier, Guillaume Peyronnet and Myself. I did some modifications, so I endorse all mistakes of this version. In the last years, the way people interact with each others on the Web has drastically changed. Web sites now provide information which is an aggregation of user-generated content, generally filtered using social recommendation methods to suggest relevant documents to users.  The most known example of such a website is Digg. This is a social news website: people share content they found on the web through the Digg […]

My name is Sylvain Peyronnet. I am an Assistant Professor of Computer Science at Paris-Sud University since 2007. I am doing my research at LRI (Laboratoire de Recherche en Informatique – Computer Science Lab in English) and my teaching at IFIPS (school of engineering of the university). I joined the university after a period of 4 years in various positions: postdoc, assistant professor in a private university, entrepreneur. I earn a Ph.D. in computer science from Paris-Sud University in 2003 (yes, I came back to the same university after 4 years). I am broadly interested in applications of probabilistic and […]

Je vous ai déjà parlé de spotrank.fr, le site web collaboratif de type Digg-like qui a la particularité d’utiliser un algorithme antispam que nous avons mis au point. Quand je dis nous je parle de la petite équipe composée de Thomas Largillier, Guillaume Peyronnet et moi-même. Les sites web sociaux fonctionnent en agrégeant du contenu fourni par les utilisateurs. Ce contenu est généralement filtré, que ce soit par une méthode de filtrage collaboratif (type recommandation par Netflix), ou par une méthode de filtrage social. Digg est sans doute le meilleur exemple d’un tel site, mais la France n’est pas en […]

C’est les vacances, et pendant les vacances le blog est calme, je vous mets quand même ici une copie du mini communiqué pour la sortie de SpotRank : Spotrank.fr est un site web collaboratif de type Digg-like qui a la particularité d’utiliser un algorithme antispam élaboré par des chercheurs afin de proposer les meilleurs résultats possibles. Les sites web sociaux fonctionnent en agrégeant du contenu fourni par les utilisateurs. Ce contenu est généralement filtré, que ce soit par une méthode de filtrage collaboratif (type recommandation par Netflix), ou par une méthode de filtrage social. Digg est sans doute le meilleur […]

Mes lecteurs les plus fidèles savent déjà tout ou presque du pagerank (voir ici et ici). A quoi sert le PageRank ? A classer les pages d’une manière qui serait cohérente avec le comportement naturel d’un internaute moyen, qu’on appelle le surfeur aléatoire, que j’appelle l’internaute crétin : il parcourt le web en suivant les liens au hasard, sauf dans certains cas où il se téléporte sur une page choisie au hasard au sein de toutes les pages présentes sur le web. Naturellement, on comprend que cette hypothèse de comportement totalement aléatoire n’a que peu de chance de correspondre au […]