• June 2020
    M T W T F S S

Building Trustworthy Big Data Algorithms

Northwestern University Newscenter (01/29/15) Emily Ayshford

Northwestern University researchers recently tested latent Dirichlet allocation, which is one of the leading big data algorithms for finding related topics within unstructured text, and found it was neither as accurate nor reproducible as a leading topic modeling algorithm should be. Therefore, the researchers developed a new topic modeling algorithm they say has shown very high accuracy and reproducibility during tests. The algorithm, called TopicMapping, begins by preprocessing data to replace words with their stem. It then builds a network of connecting words and identifies a “community” of related words. The researchers found TopicMapping was able to perfectly separate the documents according to language and was able to reproduce its results. Northwestern professor Luis Amaral says the results show the need for more testing of big data algorithms and more research into making them more accurate and reproducible. “Companies that make products must show that their products work,” Amaral says. “They must be certified. There is no such case for algorithms. We have a lot of uninformed consumers of big data algorithms that are using tools that haven’t been tested for reproducibility and accuracy.”


Boosting Productivity and Inspiring Innovation Through the Cloud

CORDIS News (08/05/13)

The European Union-funded fostering mobile business through enhanced cloud solutions (MO-BIZZ) project aims to demonstrate the importance of mobile cloud computing to businesses by encouraging small and medium enterprises to deliver business solutions that fully exploit the new technology.  The project’s researchers say MO-BIZZ will help businesses that want to access mobile network assets, such as user billing, short-messaging service, and user location.  Mobile cloud computing combines mobile computing, cloud computing, and wireless networks, which can enhance the capabilities of mobile devices and provide a much richer user experience.  The researchers say MO-BIZZ uses the resources of various clouds and network technologies to provide unrestricted functionality, storage, and mobility.  Over the next three years, MO-BIZZ plans to develop several pilot projects to showcase the potential of mobile cloud computing as a productivity booster for companies.  The project also will build a strategic global approach to the mobile cloud and develop international cooperation in order to create a thriving community of developers.  The researchers note that app developers and technology providers who want to be able to easily deploy their business apps and access mobile network assets stand to benefit the most from this technology.


Moving Into the Cloud

CORDIS News (02/14/13)

The European Union’s MobiCloud project aims to create an online collaborative platform using cloud technology to facilitate the development of mobile applications for public transport, construction, and other business-critical areas. End users, mobile developers, application vendors, system integrators, and cloud service providers will come together on MobiCloud to create end-to-end solutions with a large return on investment. MobiCloud offers a mobile mash-up screen that culls data from different corporate information technology systems, and displays different services based on a user’s context, such as location or skill set. These services respond in real time to changes such as work orders or fault reports. MobiCloud will enable smaller companies to quickly develop mobile versions of existing applications, reduce costs, and increase economic growth. The project is co-funded by the European Commission under the ICT Policy Support Program, part of the Competitiveness and Innovation Framework Program.


A Bandwidth Breakthrough

Technology Review (10/23/12) David Talbot

Academic researchers have developed coded Transmission Control Protocol, a method for improving wireless bandwidth by one order of magnitude that involves using algebra to overcome the network-clogging task of resending dropped packets.  The technology provides new ways for mobile devices to solve for missing data, and can combine data streams from Wi-Fi and LTE.  Massachusetts Institute of Technology researchers tested the system on standard Wi-Fi networks, where 2 percent of packets are normally lost, and found that a normal bandwidth of 1 Mbps was boosted to 16 Mbps.  The researchers also tested the technology in the Amazon cloud.  Internet Protocol traffic was sent to Amazon, encoded, and then decoded as an application on phones.  The technology sends algebraic equations that describe a series of packets.  That way, if a packet is lost, the receiving device can solve for the missing packet instead of asking the network to resend it.  If the technology works in large-scale deployments as the researchers expect, it could help delay a spectrum bottleneck.


Harness Unused Smartphone Power for a Computing Boost

New Scientist (08/29/12) Jacob Aron

Technical University of Braunschweig researchers have found that smartphones can be joined together in a network, which when connected via Wi-Fi, can carry out increased numbers of megaflops.  The researchers joined six low-powered phones and found they could carry out a combined 26.2 million calculations per second.  Although that performance figure is low when compared to the processing power of a modern desktop computer, the research suggests that larger smartphone clusters could be useful.  The system would be most powerful when there are large groups of phones charging at the same time.  “The more people show up, the more computer power you potentially have available,” says University of Bristol researcher Simon McIntosh-Smith.  A business model could be developed to provide incentives for users to join, such as receiving subsidized phones for users who contribute time to the cluster, says Braunschweig researcher Felix Busching.