Advancements in technological developments have shifted the focus from risk mitigation and prevention to threat prediction and aversion while delivering a delightful customer experience. Achieving both security and user experience has cast data science into the spotlight.
In 2022, artificial intelligence (AI) and machine learning (ML) will become more tightly intertwined into the fabric of trusted identity solutions across the physical and digital continuum, leading to the automation and optimisation of performance, accuracy, safety and security. This paves the way for the next frontier in physical and logical access security based on behavioural patterns and predicting anomalies.
By capitalising on the increasing amount of data generated across devices and access points, the security industry will benefit tremendously from using digital information to enhance security operations without adding too much friction for users.
According to the International Data Corp (IDC), Asia Pacific perceives the adoption of AI and automation technologies as strategic opportunities and has invested comprehensively to improve capabilities. Ninety per cent of enterprises in the region are projected to merge human expertise with AI, ML and pattern recognition to reinforce foresight across the organisation by 2026, moulding more productive and skilled workers by 30%.
Harnessing data science for protection
The integration of the Internet of Things (IoT), cloud and mobile technologies is steadily driving the digital transformation of the future. While this technological wave is creating exciting new opportunities, it is simultaneously fuelling both physical and digital security threats.
See also: Keys to achieving human-centred automation testing
As applications utilising AI and ML technologies proliferate throughout the security infrastructure, organisations can gain insights about customers’ usage of their physical and digital assets and detect abnormalities, to prevent fraudulent actions while reducing user interventions. This is achieved by analysing growing volumes of data from security devices and systems.
Additionally, AI and ML can enable buildings to be more data-driven and user-centric, making systems more manageable through a combination of physical and digital credentials and connected IoT endpoints.
Best practices
See also: Human element still important for effective mass communication
Data science offers security professionals a pattern of understanding that facilitates personalisation, eliminates friction and provides seamless services dynamically.
The following are some best practices to help enterprises leverage data science to improve their security position:
- Devise a clear data strategy – Raw data is not entirely useful. In order to gain insights from data at scale, it is imperative to have a data strategy in place. Accounting for several aspects within that strategy include the framework and tools and their relevant applications. As the framework and tools are generic, the applications are required to be highly contextual to the final outcome.
- Incorporate data management – The best-case scenario is that data is sent from a single system, formatted properly and ready for analysis. The reality is that the quality of data varies widely.
Data cleaning will almost always be necessary, but accounting for different characteristics will help collect usable data, which can then be cleaned and organised. After all, valuable insights gained can only be as good as the quality of the data. - Proactively extend AI and ML technologies – Tightly integrating these capabilities into unified physical and digital security systems will ensure a rapid response for the most successful threat detection, prediction and mitigation. By extending AI and ML technologies into the physical security world, the enterprise security operation transitions from a reactive to a proactive role in managing business operations, growth and safety.
AI algorithms could synthesise and correlate physical security information from many sources, devices and systems to create a complete view of employees, contractors and visitors. This will yield more accurate security decisions and enable organisations to better support dynamic business environments. - Automate decision making – Threats can not only be predicted and averted using data science, AI and ML can also help enterprises automate a variety of decision-making processes – from optimising business processes to proactively identifying risks and automating preventative actions – thereby improving overall organisational efficiency and effectiveness.
- User-centric system – While it is easy to think of enhancing security at the expense of user experience, data science can help overcome the challenge. AI and ML can help gain a better understanding of contextual normal usage and eliminate friction during genuine usage. To achieve this, it is important to design the system by keeping the user’s primary interaction in mind.
Ramesh Songukrishnasamy is the senior vice president and CTO of HID Global
