Qing Wang

Florida International University


Lecture Information:
  • March 19, 2018
  • 12:00 PM
  • ECS 349
Photo of Qing Wang

Speaker Bio

Qing Wang is a Ph.D. candidate at Florida International University’s School of Computing and Information Sciences under the supervision of Dr. Sitharama S. Iyengar. Her preliminary research was done under the supervision of Dr. Tao Li. Qing entered the Ph.D. program in Fall 2014, immediately after obtaining her Master’s and Bachelor’s degrees in Computer Science from Xidian University and Zhengzhou University. Her research focuses on developing intelligent data mining techniques for interactive recommender systems and automatic service management. Qing’s research work has been published in top conferences and journals such as ACM SIGKDD, ACM CIKM 2016, SDM, and KAIS. She also received the Best Student Paper Award from IEEE SCC 2017 and Student Travel Award from ACM SIGKDD 2017. In summer 2016 and 2017, Qing was a graduate research intern at the IBM Thomas J. Watson Research Center.

Description

Driven by the rapid changes in economic environment, business enterprises constantly explore innovative ways of expanding their outreach and gaining competitive advantage in the marketplace. Value-creating activities cannot be accomplished without solid and continuous delivery of IT services in this increasingly complex and specialized world. Services providers are expected to focus on innovation and assisting customers in their core business areas. Besides, time that domain experts spend on fixing operational issues has to be minimized. For these purposes, service providers seek to employ intelligent data mining techniques for maximizing the automation of subroutine procedures such as problem detection, determination, and resolution of the service infrastructure, which is an ultimate goal of IT service management (ITMS).

Specifically, the following issues will be studied for ITMS optimization. (1) How do we efficiently extract useful domain phrased to construct the domain knowledge base for cognitive ITMS? (2) How do we intelligently determine the ticket problem and adaptively recommend the best matching automation (i.e., scripted resolution) given the hierarchical information in IT automation services (ITAS)? (3) How do we interactively recommend a proper automation with no explicit hierarchical information and, in the worst case, with no contextual information of the incident ticket in ITAS, while still capturing the dependencies among automation as well?

My dissertation will address these challenges mentioned above. Concretely, I will focus on designing and developing novel data-driven intelligent solutions to maximize the automation of routine procedures, and thus alleviate the human efforts involved in ITMS, including (1) constructing the domain knowledge base for improving the efficiency of the resolution recommendation; (2) an online learning approach, hierarchical multi-armed bandit for context-based automation recommendation; (3) an online interactive collaborative filtering model for context-free automation recommendation.