|
Title:
|
A MACHINE LEARNING APPROACH TO FORECAST THE USAGE OF ANALYTICAL SERVICES |
|
Author(s):
|
Laura Gerlach, Dragan Milosevic and Vera G. Meister |
|
ISBN:
|
978-989-8533-80-7 |
|
Editors:
|
Ajith P. Abraham, Jörg Roth and Guo Chao Peng |
|
Year:
|
2018 |
|
Edition:
|
Single |
|
Keywords:
|
Data Mining, Clustering, Machine Learning, Prediction, Analytical Services |
|
Type:
|
Reflection Paper |
|
First Page:
|
241 |
|
Last Page:
|
245 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The need to store and analyze ever-increasing amounts of data has motivated us to design high performing analytical services. To overcome the challenge and need for enormous hardware resources, we used machine learning methods to forecast the need for future analytical services based on historical usage. This forecast will enable us to execute queries and cache results before they are actually submitted allowing us to prepare the analytical services for the upcoming user queries. Initial experiments show that both, a 77% accuracy in predicting future queries and a response time of 1 second, can be achieved. |
|
|
|
|
|
|
