Agnik's products are developed based on a strong technical
foundation of algorithms and systems for data analysis and mining in
mobile and distributed environments. Its proprietary technology is
developed based on years of research and development by its founders
and technical team members. The technical team of Agnik Research won
many prestigious awards. For example, Agnik's president is a Fellow
of the IEEE, winner of IBM Innovation Award, National Science
Foundation CAREER Award. Agnik received 2010 Frost & Sullivan
Enabling Technology of the Year Award and the 2010 IEEE Top-10 Data
Mining Case Studies Award for its work on MineFleet®
|Advanced Eigen-analysis of
Vehicle Performance Data
Onboard Vehicle Performance & Location Data Analytics
in-vehicle data analytics technology is based on the following
analysis of vehicle performance and location data: This allows
Agnik's vehicle data analytic products to perform
extensive statistical modeling and analysis with minimal
wireless data transmission cost. Unlike conventional
telematic systems that rely upon sending the raw data to
the server resulting high cost of wireless data
transmission, Agnik's products enjoys the benefit of
minimal wireless data transmission cost since sending
data patterns over the wireless network instead of the
raw data dramatically reduces the cost of data
transmission, often more than 100 times. Agniik's
Onboard Vehicle performance data mining technology is
Advanced predictive modeling and data mining technology for
analyzing vehicle data: Analyzing vehicle performance
and location data onboard often requires high performance data
stream management, machine learning, and data mining
algorithms in order to automatically learn predictive models,
classify outlier events, compute Bayesian statistics and
performing many other types of analysis that Agnik's vehicle
performance and location analytics products offer.
Agnik's products also exploit a collection of data mining
technology for distributed and ubiquitous environments. Some of
them are listed below.
Data Analytics for in-Vehicle Smart Phones and Other Mobile
Mobile devices such as cell-phones, PDAs, laptops, smart cards,
and wearable computers are increasingly being used for data
intensive applications. In some cases these devices themselves are
connected to different types of sensors that generate lots of data
(often continuous data streams). Sometimes, these devices are used
to remotely monitor data sources over a wireless network. The new
generation of data intensive applications in mobile devices needs
support for data analysis in environments with limited resources
(e.g. memory, processing power, battery power, user interface).
Agnik offers a comprehensive technical solution for data analysis
and mining in mobile environments.
Some of our core technical capabilities include the followings:
- Fast, scalable analysis of high volume data stream using
light-weight computing devices.
- Minimizing expensive communication load in data analysis over
wired and wireless networks.
- Design and implementation of data analysis techniques that
reduce battery power consumption.
- Minimizing the "footprint" of the system in order to be able
to run it in light-weight devices.
Scale Distributed Data Mining:
Agnik is spearheading the next generation distributed data mining
applications that are designed to run in a network of desktop
computers and high-end devices such as the Internet, grids, and
peer-to-peer systems. Agnik offers complete end-to-end distributed
data mining technology that allows analyzing distributed data,
comparing and aggregating results without having to centralize all
the data to a single central place.
Preserving Crypto for Data Analysis and Mining:
Can you analyze encrypted data so that certain underlying patterns
can be detected but the data is hidden from the outside world?
Agnik's proprietary Pattern Preserving Crypto (PPC) technology
offers exactly that. United States Department of Homeland Security
is using this technology for developing the next generation of
surveillance systems that monitor threats against the cyber
infrastructure of the country while preserving the privacy of the
monitored data.This technology can be used in surveillance systems
that require continuous monitoring of sensitive data, comparing
observations, and finding matches across multiple, proprietary
data sources. Library transactions, employee behavior, healthcare
data, financial data, law enforcement data and mobile
communications are just a few of the data types being monitored
today for corporate security, cyber-threat, and counter-terrorism
related tasks. Agnik's pattern preserving crypto technology will
allow detecting the threats while protecting the privacy of
law-abiding individuals, since raw surveillance data is often
divulged during the process of monitoring.
Resource-Constrained Data Analysis in Embedded and Sensor Systems:
New generation of computing platforms such as embedded devices and
sensor networks are increasingly developing applications that
generate lots of data in ubiquitous environments. Personalization,
context aware applications, and surveillance applications for such
environments require serious analysis of the real-time data
streams. However, the resource-constrained environments in these
devices offer many challenges for real-time analysis of the data.
Agnik has developed proprietary technology for real-time data
management and mining of un/semi and structured data streams.
product-line designed for onboard data stream mining in vehicles
is bringing the benefits of this technology to the commercial
fleet management industry.
For more information about the work performed in the above areas
by one of the co-founders of Agnik, please follow http://www.cs.umbc.edu/~hillol