Creating a predictive future with big data
Top HKUST computer scientist reveals extraordinary power of Web phenomenon and its applications in different industries
Recent years have seen a rapid increase in heterogeneous data over the Web of an unexpected scale and complexity. Such big data is everywhere, generated from all kinds of sources and applications such as social network services and cloud services.
Big data – is it just a marketing buzzword, a technical term – or does it carry a more important meaning?
“Big data will be a game-changer as we move further into the future,” says Qiang Yang, Hong Kong University of Science and Technology’s (HKUST) chair professor and head of computer science and engineering. “It creates new possibilities in many walks of life, all because of its predictability.”
In the healthcare industry, for example, big data predictive analytics is just about everywhere. From flagging drug interactions to predicting sepsis, from modelling emergency department use to triggering an automated phone call for a mammogram reminder, healthcare providers are leveraging patient data for an astounding array of care tasks.
The emergence of big data also presents massive opportunities for investors in the stock market, enabling traders to identify the best companies in which to invest. Big data technologies are also harnessed by internet giants such as Amazon, Alibaba and Taobao to read customers’ minds in order to tweak marketing campaign offers, refine operations, minimise risks and engage new customers.
“The magic of big data analytics lies in its ability to provide actionable insights for tomorrow based on today’s figures,” says Yang. “While big data is always discussed based on the 4Vs – volume, variety, velocity and veracity – its true value goes beyond these aspects.”
Yang suggests that a larger focus should be put on the business values of big data, its real-time nature and the integration of different data to create new opportunities.
Illustrating how creativity and imagination are essential for getting the most from big data analytics, Yang cites several interesting examples.
“Imagine that traffic jams can be mitigated by forecasting the number of cars on the road at different times of the day, that firemen can detect false alarms, that teachers can identify the most difficult chapters for students by tracking the most frequently visited sections on the online learning platform,” he says.
However, while most cities around the world are already unlocking the power of big data to gain competitive advantages, Hong Kong has only just embarked on this journey. “Despite the availability of basic technologies, we have yet to see the widespread application of big data analytics here,” Yang stresses.
If the main bottleneck is not in technology, what else can be done to step up big data momentum in the territory? The answer is “talents”.
“While market competition and business globalisation create huge demand for big data processing, we need the right skills to turn theory in practice,” says Yang, emphasising that big data analytics is an interdisciplinary subject which not only requires computer sciences, but also competence in maths, statistics and specific areas such as civil engineering and healthcare.
In order to address the challenges related to big data, and develop edge cutting technologies for deriving the maximum benefit while reinforcing its vision of preparing students to lead the world from Asia, the HKUST computer science and engineering department is launching two new courses with a special focus on big data technologies in the 2016-17 school year, including a master course and a minor programme. Candidates with good computer science working experience can apply for the master course. Both courses include internships.
It is hardly surprising that Hong Kong has a large demand for big data talents. For example, the Hong Kong Jockey Club has already asked to employ graduates of HKUST’s new big data courses. All in all, graduates of the new courses should have no worries about their career path.
“As a major organisation in Hong Kong, the Jockey Club posts a huge demand for big data analytics,” Yang says. “Internally, they have to analyse customer data to compile optimal customer services programmes. Big data analytics is also essential for maintaining horses in good health.”
Yang adds: “Externally, big data analytics offers many benefits to Hong Kong Jockey Club’s charity campaigns. For example, it can identify those elderly people with the most urgent needs, their location and living conditions. Big data analytics also helps the Jockey Club to better manage its voluntary task force so as to ensure the best division of labour and allocation of resources.”
Some people believe that big data analytics is only required by large enterprises and multinational companies. In view of the fact that more than 90 per cent of enterprises in Hong Kong are small and medium-sized businesses, many wonder how the best use can be made of it here. However, Yang emphasises that all types of companies should embrace big data technologies.
He also says that entertainment companies can use big data analytics to identify the most popular types of TV programmes and explore the best direction for future shows. Different government departments also have their own specific demands for big data analytics.
Yang has been involved in much big data-related research, with the focus on integrating different types such as text and graphics to transfer knowledge from one domain to another, as well as finding ever-more innovative ways to use it.
Furthermore, he was also involved in the joint project by HKUST and Digital China to foster the development of smart city in Hong Kong and mainland China based on big data technologies. He also played a key role in an earlier joint project involving HKUST and French-based Thales to develop big data technologies for smart cities with a focus on transportation and security. HKUST is the sole university partner in Hong Kong in this initiative.
“Big data will make the future work for us all if we continue seizing its strategic possibilities,” Yang says.