Regensburg 2013 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

AGjDPG: Arbeitsgruppe junge DPG

AGjDPG 4: Big Data (joint with SOE)

AGjDPG 4.6: Topical Talk

Thursday, March 14, 2013, 11:30–12:00, H37

Information Retrieval, Applied Statistics and Mathematics on BigData — •Romeo Kienzler — IBM Innovation Center Zurich, Switzerland

Although the majority of algorithms used for BigData Analytics have been developed decades ago, their application on BigData currently experiences a renaissance. In this talk a selection of algorithms and their application will be discussed in the context of BigData. A set of selected Use Cases in the field of Social Network Analysis, Bioinformatics, Financial Fraud Detection and Information Retrival will be discussed. Besides theoretical viewpoints this talk covers also runtime environments for BigData application and explains concepts of data parallelism, partition skew, aggregated storage to CPU bandwidth and fault tolerance on commodity hardware. Besides the omnipresent MapReduce/Hadoop example, which seems to be the de facto standard, we will also discuss massive parallel data warehousing and stream computing. Finally, a technical outlook tries to separate theory from reality, future from presence and hype from vision.