University of Ottawa - Carleton University
Ottawa-Carleton Institute for Computer Science (OCICS) Presentation
|
November 30, 2012 @ 10:00a.m.
A new Handoff Scheme for Vehicular Ad-hoc Network
|
Speaker: Antwi Daniel
Location: LMX 360 (Lamoureux) |
ABSTRACT
Organizations own data sources that contain millions, billions or even trillions of rows and these data are usually highly dimensional in nature. Typically, these raw datasets are comprised of numerous independent data sources that are too big to be copied or joined, with the consequence that aggregations become highly problematic. Consider in this scenario, a business manager who wants to answer the crucial location-related business question. âÂÂWhy are my sales declining at location XâÂÂ? This manager wants fast, unambiguous location-aware answers to his queries. He requires access to only the smallest yet relevant portion of the data as well as only attributes that correlate with the managerâÂÂs request. Data cube has been playing an essential role in fast OLAP in many multi-dimensional data warehouses. However current approaches compute the cube from the entire data and attribute space making it impractical for even state-of-the-art approaches to compute the cube from such huge data due to the exponential computation and storage complexity. Our goal is therefore is to compute a personalized cube for fast and scalable access to data. Here, we seek to compute the cube from only the most relevant portion of a massive data set for a given user. Our aim is first to identify the smallest data subset and its corresponding attributes sufficient for all user queries; thus ensuring that query response is fast and unambiguous. Secondly we aim to propose a new algorithm that will reduce the storage and computation complexity of current approaches. Finally, we consider the fact that users move over time and also user behaviour changes; therefore detect changes in user behaviour by constantly learning usage pattern. |
| Return to Schedule |
|
