August Research Picks

A cyber storm in a key map

Computer scientists in the USA have devised a method for analyzing, visualizing and sharing abhorrent network activity that occurs during an internet security breach or cyber storm. The maps, which utilize a metaphor based on familiar weather pattern maps, can help systems administrators and others unlock the source of a particular attack, block off access to vulnerable computers while security checks are made and defenses put in place. The researchers describe their mapping approach to such cyber storms as having the potential to mitigate attacks proactively. The same approach might be rolled out to consumers to help them better protect their personal computers when accessing the internet too.

Ferebee, D., Dasgupta, D. and Wu, Q. (2014) ‘Information assurance: a cyber security storm map’, Int. J. Information Privacy, Security and Integrity, Vol. 2, No. 1, pp.37–55.

Hospitality and terrorism

By definition terrorism strikes terror directly into those affected and more widely by generating fear and anxiety among the population. As the attack on Boston, during the city’s running marathon of 15th April 2013 demonstrated the element of surprise can increase fear and panic significantly. A collaboration between researchers in Argentina and the USA suggests that large-scale public events are particular vulnerable. Terrorism represents a serious problem for the tourist industry, although the precise risk is highly dependent on the local economy. The team suggests that where a poor region launches a large public event it can boost morale but also raise resentment among the disaffected increasing the risk of terrorist activity around the event. The team also points out that many of those involved in terrorism were raised and educated in the West and they often apply the management and marketing strategies of the West with tragic consequences for their victims. With these two concepts in mind, the team says events managers and organizers must revise their approach to risk assessment around large-scale public events.

Skoll, G.R. and Korstanje, M.E. (2014) ‘Terrorism, homeland safety and event management’, Int. J. Hospitality and Event Management, Vol. 1, No. 1, pp.95–110.

Spotting that one bad apple

Fruit disease can devastate crops and ruin economies. Conventionally, the detection of that “one bad apple” is done visually by fruit pickers, packers or others involved in fruit production. Unfortunately, there are many different afflictions that can affect a plant and the fruit it bears that are not always detected in a timely manner. Now, researchers in India are developing fruit disease recognition software that could be used to scan a batch for early signs of fungal and other infections that might otherwise be overlooked by human checkers. The software assesses surface texture and in a proof of principle demonstration, the team reports 97% accuracy on checking apples. They have also demonstrated 99.9% accuracy with the addition of gradient filters to the camera used to scan the apples.

Dubey, S.R. and Jalal, A.S. (2014) ‘Fruit disease recognition using improved sum and difference histogram from images’, Int. J. Applied Pattern Recognition, Vol. 1, No. 2, pp.199–220.

Losing big data in the cloud

A collaboration between computer scientists in Italy and Poland has investigated the problems facing those working with so-called “big data” in a cloud computing environment. There is currently explosive growth in the gathering, processing and storage of large quantities of data. This data might include military intelligence, government statistics on public behavior, consumer activity and public medical records. As such, there is a growing need for increased computer power and higher-capacity servers to handle and make use of this data. The advent of cloud computing in which storage, processing and access to data are carried out on remote and often distributed computer networks rather than the end-user’s computing facilities, whether personal or institutional, holds great opportunities for big data. Concomitantly, security, privacy and integrity of all the very different types of data held in the cloud poses new problems for the owners of that data. The team’s survey and assessment of current offerings suggests that the immature arena of cloud computing does not yet have the capacity nor the infrastructure to address all the concerns regarding big data.

Di Martino, B., Aversa, R., Cretella, G., Esposito, A. and Kołodziej, J. (2014) ‘Big data (lost) in the cloud’, Int. J. Big Data Intelligence, Vol. 1, Nos. 1/2, pp.3–17.

Author: David Bradley

Award-winning, freelance science writer based in Cambridge, England.