Tuesday, August 9, 2011

Internship: Machine Learning/Artificial Intelligence Application Intern

Company: Intel Corporation
 
Contact: Rajah Subramanian (rajah.a.subramanian@intel.com)
 
Location: Phoenix, AZ
 
Duration: September 2011 – December 2011 (can be extended full or part time based on necessity and mutual agreement)
 
Project Description:
 
This project is to explore artificial intelligence techniques to improve and automate problem management analysis to aid the problem specialists get to the root cause or problems faster. We have over a million incidents in our environments and a manual analysis of these incidents is close to impossible. Using AI we expect to understand the “clusters” and the “possible causations” from a data centric perspective and enable the problem management process to be successful.
 
The project involves the following tasks:
 
  1. Working with stakeholders to identify the correct data set from various data sources.
  2. Pull the data from these sources and perform necessary manipulation to prepare the data set for machine learning analysis.
  3. Work with subject matter experts to build data models for specific problems identified in the proof of concept.
  4. Build/Recommend GUI for displaying results from the machine learning analysis.
  5. Present results to stakeholders in a timely manner and constantly work with them for feedback and improvements.
  6. Build solutions that are reusable and re-implementable – the proof of concept will eventually be an IT wide application for data analysis.
  7. Potentially publish or present results in industry/research conferences. 
Education:
                Pursuing M.S or PhD in USA.
                (PhD preferred)
 
Skills Required:
Data Mining,
Machine Learning,
Artificial Intelligence,
Cluster Analysis,
Statistical Tools with Visual Representation,
Database Management,
SQL or other Query languages,
Good programming skills in any language – for pulling data from various sources.
Students with previous experience working with large data sets are preferred.
 

No comments:

Post a Comment