
Supports de cours -- Data Mining Cette page recense les supports utilisés pour mes enseignements de Machine Learning, Data Mining et de Data Science au sein du Département Informatique et Statistique (DIS) de l'Université Lyon 2, principalement en Master 2 Statistique et Informatique pour la Science des donnéEs (SISE), formation en data science, dans le cadre du traitement statistique des données et de la valorisation des big data. Je suis très attentif à la synergie forte entre l'informatique et les statistiques dans ce diplôme, ce sont là les piliers essentiels du métier de data scientist. Attention, pour la majorité, il s'agit de « slides » imprimés en PDF, donc très peu formalisés, ils mettent avant tout l'accent sur le fil directeur du domaine étudié et recensent les points importants. Cette page est bien entendu ouverte à tous les statisticiens, data miner et data scientist, étudiants ou pas, de l'Université Lyon 2 ou d'ailleurs. Nous vous remercions par avance. Ricco Rakotomalala – Université Lyon 2
Top 10 data mining algorithms in plain English Today, I’m going to explain in plain English the top 10 most influential data mining algorithms as voted on by 3 separate panels in this survey paper. Once you know what they are, how they work, what they do and where you can find them, my hope is you’ll have this blog post as a springboard to learn even more about data mining. What are we waiting for? Let’s get started! Update 16-May-2015: Thanks to Yuval Merhav and Oliver Keyes for their suggestions which I’ve incorporated into the post. Update 28-May-2015: Thanks to Dan Steinberg (yes, the CART expert!) What does it do? Wait, what’s a classifier? What’s an example of this? Now: Given these attributes, we want to predict whether the patient will get cancer. And here’s the deal: Using a set of patient attributes and the patient’s corresponding class, C4.5 constructs a decision tree that can predict the class for new patients based on their attributes. Cool, so what’s a decision tree? The bottomline is: Is this supervised or unsupervised? 3.
Sipina - Arbres de décision Scheduling in Hadoop Hadoop is a general-purpose system that enables high-performance processing of data over a set of distributed nodes. But within this definition is the fact that Hadoop is a multi-tasking system that can process multiple data sets for multiple jobs for multiple users at the same time. This capability of multi-processing means that Hadoop has the opportunity to more optimally map jobs to resources in a way that optimizes their use. Up until 2008, Hadoop supported a single scheduler that was intermixed with the JobTracker logic. Luckily, a bug report (HADOOP-3412) was submitted for an implementation of a scheduler that was independent of the JobTracker. With this change, Hadoop is now a multi-user data warehouse that supports a variety of different types of processing jobs, with a pluggable scheduler framework providing greater control. Note: This article assumes some knowledge of Hadoop. The core Hadoop architecture Figure 1. Back to top Hadoop schedulers FIFO scheduler Fair scheduler
Web Squared Journal Observations About Streaming Data Analytics for Science | The eScience Cloud I recently had the pleasure of attending two excellent workshops on the topic of streaming data analytics and science. A goal of the workshops was to understand the state of the art of “big data” streaming applications in scientific research and, if possible, identify common themes and challenges. Called Stream2015 and Stream2016, these meetings were organized by Geoffrey Fox, Lavanya Ramakrishnan and Shantenu Jha. First it is important to understand what we mean by streaming data analytics and why it has become so important. In some cases, the volume and rate of generation is so large, we cannot keep the data at rest for very long. This article has two parts. There are many factors that determine when a particular technology is appropriate for a particular problem. We can divide the spectrum of streaming data scenarios into three basic categories The data streaming challenges that confront large enterprises when dealing the data from millions of users of Internet enabled devices.
actuvisu Blog Postgres-XL | Open Source Scalable SQL Database Cluster Datavisualization MongoDB Performance Tuning: Everything You Need to Know MongoDB is one of the most popular document databases. It’s the M in the MEAN stack (MongoDB, Express, Angular, and Node.js). Unlike relational databases such as MySQL or PostgreSQL, MongoDB uses JSON-like documents for storing data. MongoDB is free, open-source, and incredibly performant. However, just as with any other database, certain issues can cost MongoDB its edge and drag it down. In this article, we’ll look at a few key metrics and what they mean for MongoDB performance. Performance of locking in transactionsMemory usageConnection handlingIssues with replica sets Of course, MongoDB performance is a huge topic encompassing many areas of system activity. Now, let’s get into it. Analyze locking performance How does MongoDB handle locking? For example, if a client attempts to read a document that another client is updating, conflicts can occur. When a lock occurs, no other operation can read or modify the data until the operation that initiated the lock is finished. locks.
Blog eye tracking : ergonomie web, tests utilisateur | Pertech Le eye tracking comme outil pour la sécurité aérienne Le magazine belge L’écho publie aujourd’hui un article sur la technologie de eye tracking Pertech. Cet article présente l’oculomètre Pertech entre comme outil pour la sécurité aérienne. L’article est à télécharger au format pdf ici-même : Le eye tracking comme outil pour la sécurité aérienne Nouvelle version du site Pertech Notre site bénéficie d’une toute nouvelle version. Le eye tracker Pertech pour écrire avec les yeux Et si l’écriture avec les yeux n’était plus de la science-fiction? Ces projets de recherche ont donc abouti sur un système à base de eye tracking permettant d’écrire avec les yeux. Pertech et Jérôme Baujon dans le magazine de la CCI – Juin 2012 Jérôme Baujon présente Pertech dans le magazine de la CCI Sud Alsace du mois de Juin 2012 Conférence eye tracking Pertech – Bruxelles le 17 avril 2012 Qu’est-ce que le Eye-Tracking ? Elle prend tout son sens dans la validation du comportement du sujet face à une tâche.
4 Promising Curation Tools That Help Make Sense of the Web Steven Rosenbaum is a curator, author, filmmaker and entrepreneur. He is the CEO of Magnify.net, a real-time video curation engine for publishers, brands, and websites. His book Curation Nation is slated to be published this spring by McGrawHill Business. As the volume of content swirling around the web continues to grow, we're finding ourselves drowning in a deluge of data. The solution on the horizon is curation. In the past 90 days alone, there has been an explosion of new software offerings that are the early leaders in the curation tools category. 1. Storify co-founder Burt Herman worked as a reporter for the Associated Press during a 12-year career, six of those in news management as a bureau chief and supervising correspondent. At the AP, editors sending messages to reporters asking them to do a story would regularly write, “Can u pls storify?” Storify uses existing elements from the web and gives curators the power to drag and drop elements into storylines. 2. 3. 4. Conclusion