Agnik Technology


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® technology.
technology_agnik
Advanced Eigen-analysis of Vehicle Performance Data

Onboard Vehicle Performance & Location Data Analytics

Agnik's core in-vehicle data analytics technology is based on the following foundations:

  1. Onboard 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 patent protected.
     

  2. 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 Devices:

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:
  1. Fast, scalable analysis of high volume data stream using light-weight computing devices.
  2. Minimizing expensive communication load in data analysis over wired and wireless networks.
  3. Design and implementation of data analysis techniques that reduce battery power consumption.
  4. Minimizing the "footprint" of the system in order to be able to run it in light-weight devices.


Large 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.

Pattern 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.

Real-time 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. Agnik's MineFleet® 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