Abstract:

During the last decade we have witnessed a huge growth in data, up to 2.5 quintillion bytes of data created every day so far. This data comes from a variety of sources such as postings in social media sites, digital pictures and videos, traffic information from loop sensors, GPS signals from cell phones, stock exchange prices, purchase transaction records, to name a few. Big data never sleeps, it is a continually growing stream of digital activity pulsating through wires and air across the world. But big data is more than just a matter of volume, it is an opportunity to gain actionable insights from new and emerging types of data and content, left untapped by traditional business intelligence, to help companies make better business decisions and become more agile, effective and competitive. In this talk I will present some scenarios to motivate the need for a new kind of analytics platform, Live Analytics, capable of dealing with the new challenges that big data impose along the dimensions of volume, velocity and variety to deliver insights and predictive analytics within actionable time windows (i.e., "at the speed of business"). Examples of such scenarios include situational awareness, sentiment analysis, traffic impact prediction and activity monitoring in healthcare. Solutions built on top of Live Analytics need to integrate information of diverse data types from an increasing number of sources, handle explosive growth in data volumes, deal with rapidly updated content, and deliver shorter cycle times to quality decisions as well as a higher degree of automation. I will describe our perspective of the challenges that we have faced in building such solutions and the technologies built into the Live Analytics platform to support these novel requirements.

Biography:

Professor Malu Castellanos is a senior researcher and technical lead in the Information Analytics Laboratory at Hewlett-Packard Laboratories in Palo Alto, CA, USA. Since 1997 she has been applying information management and analytics technologies to develop intelligent solutions for different kinds of business related problems and novel techniques for different aspects of business intelligence. She received a B.S. in Computer Engineering at the National University of Mexico and a Ph.D. in Computer Science from the Polytechnic University of Catalunya. Prior to joining Hewlett-Packard she was on the faculty at the Information Systems Department of the Polytechnic University of Catalunya. She has 12 granted patents, 17 pending and more than 60 publications in international conferences, journals and book chapters. She has actively participated in numerous Program Committees (including VLDB, SIGMOD and ICDE), journal review boards and advisory boards and has held different leadership roles in the organization of several IEEE and ACM international conferences such as being General Chair of IEEE ICDE 2008 and PC Chair of EDBT 2012 Industry and Applications Track. She initiated and has been organizing the workshop series Enabling Real-Time Business Intelligence (BIRTE) in conjunction with VLDB since 2006 and Business Process Intelligence (BPI) in conjunction with BPM (from 2005 until 2010). She currently serves as a member of the Executive Committee of IEEE Technical Committee of Data Engineering. Her interests are information management technologies in general, and in particular, real-time business intelligence, text analytics, and data management platforms.