My research has addressed some of the critical shortcomings of prior works in augmenting the request resolution performance of the aforementioned systems. In particular, I have proposed novel search and routing strategies that minimized the resolution time of a generated query or incident. In order to assess the performance of the proposed and alternative search strategies, I have also derived theoretical models of the communities to provide a complete representation of a commuity with details that are often missing in one data set.

During the early stage of my doctoral study, I have also contributed to research on designing energy-efficient protocols for critical applications of a sensor network such as monitoring of an unmanned terrain using mobile sensor nodes; scheduling activation/deactivation of sensor nodes to prolong life-time of a stationary sensor network; and adjusting transmission power of nodes during broadcasting to conserve energy.

Contact Information

E-Mail: EMAIL OBFUSCATED

Phone: 716-908-8746

• Data Aggregation assuming asynchronous communication – Asynchronous communication system assumes the processing time of each node (agent) is not bounded. Therefore, the agents in the network do their voting for each round at different points of time. DA with asynchronous communication is suitable for situations where the stations are mobile, or the speeds of communications links are not the same or vary. Therefore, the convergence expressions and convergence properties are being investigated for such networks.
• Multi-domain decision making process – The data aggregation problem mostly assumes the values to vote on are singular, such as receiving a single value from each neighbor and vote on the values received. However there are applications where the data received from each neighbor are multi-domained or belong to multiple domains. An example is receiving the location of an object in terms of longitude, latitude, and altitude. Therefore, the question is how to vote on these multi-domain values, one received from each neighbor. A simple application is successfully identifying the location of an object.
• Target detection – Another avenue of research currently being investigated is in reference to “target detection”. The objective is to detect a target successfully through a distributive and collaborative decision making process. There are a number of challenging items to deal with such as distinguishing between weak sensing signals and noise signals, how to mask the faulty information from neighbors, or when not all sensing devices can detect the object in order for the network to make the right decision.

Contact Information

E-Mail: EMAIL OBFUSCATED

Phone: 4022148226

E-Mail: EMAIL OBFUSCATED

Phone: +33492965182

E-Mail: EMAIL OBFUSCATED

Scientific workflows are characterized by their complex and involved dataflow patterns. In order to efficiently enact such workflows on large scale grid infrastructures, a sophisticated set of modalities are required that process asynchronous dataflows while at the same time taking advantages of the parallel execution of many jobs on Grid. In this direction, our efforts involved development of a sophisticated workflow language called GWENDIA and enactment of biomedical applications workflows on production grids. Other efforts involved development of a robust grid interface to the Taverna workbench in the form of a grid middleware plugin for the European EGEE grid from within Taverna workbench.

Scientific workflow users predominantly use the visual mode for workflow development. However, compact and portable scripts have their own advantages leading to rapid development of expressive and parallel workflows. In this direction, we exploited the scripting paradigm of workflow expression and developed a script equivalent of the GWENDIA workflow language called gscript. The gscript language is syntactically different, semantically equivalent workflow language that takes advantage of the same underlying enactment engine MOTEUR. A two way translator was developed that can translate the GWENDIA workflows to gscript and vice versa.

While the current efforts in the area of workflows for large scale infrastructures are maturing, the direction they are taking tends to be specific which results in a limited use of a particular Workflow platform and redundant developments. This motivates a need for an interoperable and widely usable workflow platform. A promising future research area could be to achieve such an interoperability among diverse workflow platforms and use the existing repositories such as myexperiment.org in order to reuse the workflow components in new ways and combinations. Such a research can significantly facilitate scientific endeavors in diverse domains and rapid experimental results taking advantages of Grids and emerging cloud and virtualization infrastructures.

Contact Information

E-Mail: EMAIL OBFUSCATED

Phone: +33492965182

Education

University of Kansas
Ph.D. in Computer Science
Expected July 2010.
Advisor: Prof. Victor S. Frost
Dissertation: “Optimal Communications Systems and Network Design for Cargo Monitoring.”

University of Missouri-Kansas City
M.S. in Computer Science with an emphasis in Networking and a secondary concentration in Telecommunications, awarded December 2005.
Advisor: Dr. Cory C. Beard
Thesis: “The Effects of MAC Layer Parameters on Quality of Service Provided by 802.11e.”

Park University
B.A. Magna Cum Laude in Computer Science with a minor in Mathematics, awarded December 2000.

Full CV is available here.

Contact Information

E-Mail: EMAIL OBFUSCATED

I also have a strong background as a software engineer. Before I came to the U.S. for pursuing Ph.D. study, I worked as a software engineer for eight years, three of those years at Samsung Electronics. As a software engineer, I developed system programs such as device drivers of Windows CE for Samsung products, as well as Windows applications. This experience enhances my research ability especially for creation and implementation of an algorithm.

Contact Information

E-Mail: EMAIL OBFUSCATED

Phone: (515) 676-0114

Research Philosophy

My research method is to collect real time network traffic and conduct analysis on that to gather useful information to understand the symptoms of network attacks and misuses. I can use these findings to design novel techniques which can detect such anomalies. I believe that experimenting with real network data is necessary to make any useful contribution although analysis of a simulated data may be a starting point. To support my thesis research I have collected and analyzed network traffic at the departmental network of my university.

Contact Information

E-Mail: EMAIL OBFUSCATED

In my most recent work I have focused on making sense of social (textual) data, drawing from the fields of statistical Natural Language Processing (NLP), Machine Learning, Social Web and the Semantic Web to tackle a diverse set of information extraction problems and to power Social Intelligence applications. My research has focused on answering three broad questions in the analysis of Social Media content:
- Characterizing what people are talking about What are the named entities and topics that people are making references to? How are cultures interpreting any situation in local contexts and supporting them in their variable observations on a social medium?
- Characterizing how they express themselves What do word usages tell us about an active population or about individual allegiances or non-conformity to group practices?
- Characterizing why they scribe What are the diverse intentions that produce the diverse content on social media? Can we understand why we share by looking at what we predominantly do with the medium? What emotions are people conveying about a person, event or incident?

Several results of my work have been absorbed into two deployed social intelligence Web applications. First, the BBC SoundIndex (www.almaden.ibm.com/cs/projects/iis/sound/), that measures the pulse of a music populace by tapping into content from online music communities for the end goals of popularity mining. The second web application, Twitris (http://twitris.knoesis.org) aggregates user perceptions behind real-time events using data from Twitter.

My future interests are in grounding social data analysis in Communication theory, understanding a social process in the context of people, content and network connections –“who (people) said what (content) to whom (network structures) in what channel with what effect”; and in interpreting results of the computational analyses using theories of information/social science.

More information can be found on my webpage http://knoesis.wright.edu/researchers/meena/ and CV http://knoesis.wright.edu/researchers/meena/homepage/resume.pdf

Contact Information

E-Mail: EMAIL OBFUSCATED

Contact Information

E-Mail: EMAIL OBFUSCATED