Harnessing the capability of AI to help resolve global challenges

Researchers at Qassim University are developing AI-based totally programmes to address a wide type of challenges, from on-line safety to healthcare
As the arena continues its rapid digital transformation, ensuring that those new technology serve and guard on line groups is a growing assignment for innovators within the computing sphere.

At Qassim University in Saudi Arabia, specialists are harnessing the capacity of synthetic intelligence (AI) and gadget mastering to broaden defences in opposition to current threats inclusive of cybercrime and the unfold of misinformation. Meanwhile, comparable AI-based programmes can be used to broaden neighborhood initiatives including agriculture and to support the ones dwelling with disabilities.

At Qassim University, AI studies is being led by means of numerous researchers, along with Suliman Alsuhibany, dean of the pc college, and Mohammad Alsuhaibani, assistant professor at the college and head of the pc technological know-how branch.

Developments in AI have expanded rapidly in recent years, and college-led studies has already furnished many answers by using way of labour automation. But as the field of research expands, it’s far Alsuhibany’s perception that AI will play an increasingly more crucial position in cybersecurity.

“An growing attention for us could be accuracy in cybersecurity programs, improving AI algorithms constantly to mitigate towards destiny attacks as far as possible,” he explains. “Right now, we are detecting malware and attacks as they appear, but the purpose in the following few years can be to installation AI to truely save you assaults from going on within the first location.”

To progress in this vicinity, Qassim researchers are working to enhance the AI algorithms themselves. Alsuhibany offers the instance of e mail filters, that may correctly installation gadget learning tools to discover and classify incoming messages as junk mail. “These algorithms are getting greater sophisticated, permitting the inbox to filter out for promotional content or advertising emails, but there is still much room for development,” he says.

Experts in cybersecurity warn that cyberattacks are becoming an increasing number of frequent and sophisticated. Protecting governments, industries and local groups towards malicious parties is critical for each national and geopolitical safety. Here, AI-based totally research is supplying answers.

“Computer networks are a target for several kinds of attack every hour of every day, and these malicious programmes are constantly evolving,” says Alsuhibany. “We can use device learning to feed what we call an thoughts intrusion detection machine,” he adds, which means that, because the cyber threats evolve to grow to be extra state-of-the-art, detection programmes evolve alongside them.

While cybersecurity will continue to be a middle consciousness of the computer college research organization, initiatives will vary extensively of their scope.

One current take a look at, co-authored by using Alsuhaibani, demonstrates how AI may be used to stumble on fabricated non secular texts, or mawdu’ hadiths – a chronic problem faced with the aid of pupils on-line. His studies crew developed a model that may determine variations in grammar and textual content patterns indicative of faux scriptures. The programme finished a ninety two.Forty seven in keeping with cent fulfillment price whilst examined in opposition to inauthentic texts. It’s a project that Alsuhaibani hopes could also have a big impact on society by decreasing the ability for non secular intolerance and extremism.

Another mission led with the aid of the university applies deep mastering and AI to agriculture – specifically, date fruit reputation. Saudi Arabia produces extra than 450 varieties of dates, meaning sorting and figuring out those fruits is a first-rate venture requiring sizeable sources. Through system studying, the group was capable of broaden algorithms to hit upon and recognise the distinctive culmination through automation, rushing up the manner for industry employees inside the busy harvesting months.

One of the college’s proudest achievements is a venture currently underneath manner in collaboration with the Saudi Arabian visually impaired network. As smartphones an increasing number of permit users to free up their monitors thru a pattern password, visually impaired users are excluded from the use of such programmes. Alsuhibany sought to make this system feasible for them by using exploiting haptics – the small vibrations smartphones make in reaction to the touch – to introduce a vibration-primarily based sample password technique.

“We attributed every variety on the display screen a completely unique vibration measured to the millisecond,” Alsuhibany explains. “Then we undertook a usability take a look at that generated some very treasured remarks from visually impaired customers. They explained to us when vibrations have been too comparable, for example, and we adapted it consequently.”

Going forward, it’s far Alsuhibany’s hope that the laptop university will have interaction in greater go-disciplinary paintings, growing AI-based totally solutions for healthcare, among different vital areas. “Establishing network-primarily based research collaborations, which includes the pattern password have a look at, might be very important for this paintings,” he says.

The university has labored on other healthcare-related projects in latest months; VacChain is a blockchain generation programme designed to help medical experts coordinate and digitise children’s vaccine records. The programme has been picked up through the government ministry of health for a ability pilot take a look at. If the pilot is a hit, it may be rolled out across the usa.

“One of the maximum not unusual concerns people have approximately trying out those technologies is defensive the privateness of the affected person,” Alsuhibany says. “But we’ve got approaches to ensure the generation is secure, using firewalls, as an example.”

The group have even been able to observe computational research practices within the combat towards Covid-19. Recognising a aid hole that meant many worldwide healthcare systems struggled to successfully treat and diagnose Covid-19 patients at pace, Qassim researchers proposed a “deep getting to know” technique to appropriately detecting infections based on chest X-ray photographs.

For the programme, named COVID-CGAN, researchers generated a huge information set the use of unique and laptop-generated snap shots. The information set became then used to train pc fashions to as it should be locate the ailment in extra than 99 consistent with cent of sufferers.

“Importantly, nearly all the research we do on the college is supported via neighborhood and international collaborations,” Alsuhaibani says. “These are demanding situations that have an effect on a extensive wide variety of human beings and places, and so we have to paintings together across disciplines to locate answers.”