Electrical & Computer Engineering Department
DATA MINING AND ANALYTICS - FROM TEXT MINING, EMAIL MARKETING EFFECTIVENESS PREDICTION TO HEALTH DATA ANALYTICS
DATE: SEPTEMBER 23, 2016 11:00-12:00AM
ROOM: SL 165 (723 WEST MICHIGAN STREET)
Computer Information and Graphic Technology
Indiana University Purdue University Indianapolis
With rapid development of electronic devices, mobile devices and computer software and applications, huge volume of data is being generated every second. Big data has hit business, government, healthcare and scientific sectors. There is no double that value, competitiveness and efficiency are driven by data in all these sectors. In this talk, I will present three projects in the area of data mining and analytics that I have worked on. These projects landed in the text mining, email marketing effectiveness prediction and health data analytics. Challenges and successes in data extraction, data representation, data mining and data analytics methodologies will be discussed. In this talk, I will also talk about the future research projects in the health informatics domain.
Dr. Xiao Luo is currently an assistant professor at the department of Computer Information and graphic technology of IUPUI. Before joining IUPUI, she is a senior data engineer at Nova Scotia Health and Wellness in Canada. She obtained her PhD of Computer Science from Dalhousie University in 2009. She previously worked on multiple Natural Sciences and Engineering Research Council of Canada (NSERC) funded projects as a Postdoctoral Fellow at Dalhousie University and Saint Mary’s University. She also worked as a research officer for National Research Council of Canada and as a research computing consultant for Atlantic Research Data Center. Dr. Xiao Luo’s research interests include intelligent and cross disciplines data analytics, applied data mining and machine learning, big data representation and integration, and automated data extraction. She has published academic papers in international conferences and industrial white papers with different organizations. She served as a PC member or reviewer for different international conferences and journals, such as Journal of Data Mining and Knowledge Discovery, Journal of Discovery Science, Recent Advances on Computational Intelligence in Defense and Security, and so on.
Archived Seminars 2016-2017
Topic Big Data Analysis Across Domains: From Recommender Systems to Chemical Informatics
Date: Friday, September 16, 2016 Time: 11:00 AM – 12:00PM
Room: SL 108 (723 West Michigan street)
Xia Ning, Ph.D.
Computer & Information Science, Indiana University – Purdue University Indianapolis
We are in the era of Big Data, where the sheer volume, high velocity, data heterogeneity and complexity have introduced unprecedented challenges to data mining and machine learning research and their applications in real life. Effectively mining, learning and eventually creating values from Big Data become critical for many high-impact application domains.
In this talk, I will address some Big-Data issues for two specific application domains, i.e., Recommender Systems and Chemical Informatics. For Recommender Systems, I will introduce SLIM, a sparse linear method, that can achieve high prediction accuracy and has low computational requirements for Top-N recommendation from Big Data. I will also present a unified frame based on SLIM that enables effective exploration and incorporation of additional meta-data for better recommendation performance. For Chemical Informatics, I will present my work iSAR on predicting compound activities, where iSAR provides a guided-search approach over the entire chemical space (~10100 compounds) by leveraging available information from related protein targets. I will also briefly talk about my work on Bioinformatics and on-going projects on drug discovery problems.
BIOGRAPHY: Dr. Xia Ning is currently an assistant professor at the Computer & Information Science Department, Indiana University – Purdue University Indianapolis (IUPUI). Before joining IUPUI in 2014, she was a researcher at NEC Labs America. She got her Ph.D. degree from the Department of Computer Science & Engineering, University of Minnesota, Twin Cities, in 2012, under the supervision of Prof. George Karypis. Dr. Ning's primary research interests lie in Big Data analytics, data mining and machine learning, with specific applications in Recommender Systems, Chemical Informatics and Health Informatics. She has published papers on both high-impact journals (e.g., Journal of Chemical Informatics and Models) and top conferences (e.g., KDD, ICDM, SDM, CIKM, Recsys, WWW and AISTATS). She has also served as a PC member for top venues as well as a panelist for NSF. Dr. Ning holds 11 pending/granted patents with NEC Labs and Qualcomm, Inc. Her research is currently supported by NSF.
Archived Seminars 2015-2016
Date: Monday, June 6, 2016
Title: Optimum Functional Forms of Spoofing Attacks, Optimum Processing for Man-in- the-Middle Attacks and Interesting Implications to Unattacked Quantized Sensor Estimation Systems
Speaker: Rick S. Blum, Robert W. Wieseman Endowed Professor of Electrical Engineering, Electrical and Computer Engineering Dept., Lehigh University
Abstract: Estimation of an unknown deterministic vector from quantized sensor data is considered in the presence of spoofing and man-in-the-middle attacks which alter the data presented to several sensors. First, asymptotically optimum processing, which identifies and categorizes the attacked sensors into different groups according to distinct types of attacks, is outlined in the face of man-in-the-middle attacks. Necessary and sufficient conditions are provided under which utilizing the attacked sensor data will lead to better estimation performance when compared to approaches where the attacked sensors are ignored. Next, necessary and sufficient conditions are provided under which spoofing attacks provide a guaranteed attack performance in terms of the Cramer-Rao Bound (CRB) regardless of the processing the estimation system employs, thus defining a highly desirable attack. Interestingly, these conditions imply that, for any such attack when the attacked sensors can be perfectly identified by the estimation system, either the Fisher Information Matrix (FIM) for jointly estimating the desired and attack parameters is singular or the attacked system is unable to improve the CRB for the desired vector parameter through this joint estimation even though the joint FIM is nonsingular. It is shown that it is always possible to construct such a highly desirable attack by properly employing an attack vector parameter having a sufficiently large dimension relative to the number of quantization levels employed, which was not observed previously. For unattacked quantized estimation systems, a general limitation on the dimension of a vector parameter which can be accurately estimated is uncovered.
Biography: Rick S. Blum received a B.S.E.E from Penn State in 1984 and an M.S./Ph.D. in EE from the University of Pennsylvania in 1987/1991. From 1984 to 1991 he was with GE Aerospace. Since 1991, he has been at Lehigh. His research interests include signal processing for smart grid, communications, sensor networking, radar and sensor processing. He was an AE for IEEE Trans. on Signal Processing and for IEEE Communications Letters. He has edited special issues for IEEE Trans. on Signal Processing, IEEE Journal of Selected Topics in Signal Processing and IEEE Journal on Selected Areas in Communications. He was a member of the SAM Technical Committee (TC) of the IEEE Signal Processing Society. He was a member of the Signal Processing for Communications TC of the IEEE Signal Processing Society and is a member of the Communications Theory TC of the IEEE Communication Society. He was on the awards Committee of the IEEE Communication Society. Dr. Blum is a Fellow of the IEEE, a former IEEE Signal Processing Society Distinguished Lecturer, an IEEE Third Millennium Medal winner, a member of Eta Kappa Nu and Sigma Xi, and holds several patents. He was awarded an ONR Young Investigator Award and an NSF Research Initiation Award.
Location:Science & Engineering Building 723 West Michigan St., SL 165 Indianapolis, IN 46202
All Seminars are held in SL 165, unless otherwise noted.