DocumentCode
2172894
Title
The eighth annual MLSP competition: Overview
Author
Montanez, Ken ; Liu, Weifeng ; Calhoun, Vince D. ; Huang, Catherine ; Hild, Kenneth E., II
Author_Institution
Amazon Corp., USA
fYear
2012
fDate
23-26 Sept. 2012
Firstpage
1
Lastpage
4
Abstract
This marks the eighth year the Machine Learning for Signal Processing (MLSP) Technical Committee has hosted a data analysis competition, which is held in conjunction with the annual MLSP workshop. For this year´s competition, which was sponsored by Amazon Corporation, entrants were asked to write an algorithm that attempts to automatically provision an employee´s access to company resources in an optimal manner. In this paper, we (the organizers of the competition) briefly describe the application, the data, the rules, and the outcomes of the competition. A total of 4 teams entered the contest. We provided real (declassified) training data to the entrants and tested the algorithms using disjoint test data. The two teams with the best performing entries describe the approach they used in two separate companion papers, both of which appear in this year´s conference proceedings.
Keywords
authorisation; business data processing; collaborative filtering; data analysis; learning (artificial intelligence); recommender systems; signal processing; Amazon Corporation; Machine Learning for Signal Processing Technical Committee; annual MLSP workshop; automatic access provisioning; collaborative filtering; company resources; conference proceedings; data analysis competition; disjoint test data; eighth annual MLSP competition; employee access; real training data; recommender systems; Companies; Conferences; Error analysis; Machine learning; Manuals; Signal processing; Signal processing algorithms; Competition; automatic access provisioning; collaborative filtering; machine learning; recommender systems;
fLanguage
English
Publisher
ieee
Conference_Titel
Machine Learning for Signal Processing (MLSP), 2012 IEEE International Workshop on
Conference_Location
Santander
ISSN
1551-2541
Print_ISBN
978-1-4673-1024-6
Electronic_ISBN
1551-2541
Type
conf
DOI
10.1109/MLSP.2012.6349769
Filename
6349769
Link To Document