A model-based OLAP approach

• Grant number: 15/11106-0
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: March 08, 2016
• End date: March 07, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Cristina Dutra de Aguiar
• Grantee: Jaqueline Joice Brito
• Supervisor: Yannis Papakonstantinou
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: University of California, San Diego (UC San Diego), United States
• Associated to the scholarship: 12/13158-9 - Storage and recovery of warehouse data in the cloud, BP.DR

Abstract

Models, and their capacity to adapt, are behind any new major technological advance we have seen in the previous decades. From maps in our cellphones to bionic limbs that bring back life quality to an increasing number of people, creating systems with underlying models that can learn and make predictions seem to be a core technique in the foreseeable future. This is directly reflected on a recent steep growth in demand for "data analysts". A common ground to all these applications is the use of models in conjunction with data (e.g., from sensors) stored in databases. For instance, Plato is a database system introduced to deal with models as a primitive data type. Therefore, the database can offer out of the box predictive abilities and a more accurate representation of the data. Nonetheless, Plato is still on an early stage of development and lacks the proper definition of complex operations. In this context, we propose the development of efficient Online Analytical Processing (OLAP) operations meeting the needs of data analysts and benefiting from several advantages of this new model-centered database. This model-based OLAP methodology should efficiently solve complex queries that may mix conventional data and models. We believe this could positively impact data analysts and serve as a general framework for Big Data Analytics. This internship project will be hosted at the University of California San Diego (UCSD) and supervised by Prof. Dr. Yannis Papakonstantinou. (AU)