Research & Projects
Research & Projects
Identifying de-facto prescription norms in a hospital setting: A study with antibiotics. IEEE International Conference on Healthcare Informatics (ICHI), Xian, China, 2019. [pdf]
UIOWAS at TREC Clinical Decision Support Track 2016, TREC Clinical Decision Support Track 2016. Gaithersburg, Maryland, November 2016. [pdf]
Enhancing SciENcv through semantic research profile integration with the VIVO-ISF ontology, VIVO Conference 2016, Denver, Colorado, September 2016. [Web link]
Malware Detection using Statistical Analysis of Byte-Level File Content, ACM Special Interest Group on Knowledge Discovery & Data Mining, CSIKDD Workshop, Paris, France, 2009. [pdf]
PE-Miner: Mining structural information to detect malicious executables in realtime, 12th International Symposium on Recent Advances in Intrusion Detection, Springer 2009. [pdf]
PE-Probe: Leveraging Morphological Information Of Portable Executables To Detect Zero-Day Malware, Virus Bulletin Conference (VB), Geneva, Switzerland, 2009. [Web link]
A Sense of Danger for Windows processes, International Conference on Artificial Immune System (ICARIS), Springer, 2009. [pdf]
On the Appropriateness of Evolutionary Rule Learning Algorithms for Malware Detection, International Workshop on Learning Classifier Systems (IWLCS), held in conjunction with Genetic and Evolutionary Computation Conference (GECCO), ACM Press, Montreal, Canada, 2009. [pdf]
Are evolutionary rule learning algorithms appropriate for malware detection? Genetic and Evolutionary Computation Conference (GECCO), ACM Press, Montreal, Canada, 2009. [pdf]
In this work, structural features are extracted automatically from headers of portable executables (PE) files and are used to classify unseen malicious PE files using data mining algorithms. Our system achieved more than 99% detection rate with less than 0.5% false alarm rate for distinguishing between the benign and malicious executables.
This project is a toolkit of statistics for multivariate data. It takes the values of multiple variables and then can perform different statistical techniques on it as required by the user. The produced result is used by statisticians to make inferences about data. The visualization of multivariate data makes the analysis simple.
It is a tool for performance analysis of various system wide features of desktop systems. It leverages the system calls, number of CPU cycles, and amount of disk space used, real and virtual memory consumption to give an overall picture of the effect of particular software on system resources
Presented a statistical model of CAIDA Internet telescope (or backscatter) traffic. Developed a statistical model of a rate-limiting detector and validated it using various packet arrival time distributions.