2018 Technology Predictions: Insights-Driven Enterprises Will Emerge as Winners in 2018

By Joseph Lee, Vice President, Asia Pacific and Japan, Kinetica

Digital insights are increasingly being used to transform the business. How fast a company can react to these insights is the difference between capturing revenue, customer sentiment or critical national safety information, or not. Within the digital transformation is the ability for a business to go from data, to insights, to action, very quickly. I was at a sold out NVIDIA Singapore AI event this year and heard how their customers are looking for the ability to consolidate, visualise and simulate multiple scenarios, and make sense of huge volumes of streaming data from within and outside of the business. This reinforced what I’ve been hearing from client visits. The need to do this in an accelerated manner is the key factor and positions companies to be much more proactive and competitive.

What we have seen in 2017

In the financial services sector, where milliseconds matter and where insights directly equate to money, organisations have turned to GPU databases to push conventional computing to its limits. By using the most-up-to-the-moment data in sub-seconds–instead of hours¬¬–organisations are able to perform risk calculations, fraud analysis and algorithmic trading in near real time.

In retail, one of Asia’s largest conglomerates, Lippo Group, is using Kinetica’s GPU database to consolidate multiple dimensions of customer attributes from their diversity of multi industries data sources, including demographics, location, and cross-channel buying behaviour such as sentiments gained from social media and interactions in physical stores and online. All generated insights are then to be encapsulated in the form of API, for any digital touch point channels to consume and then interact with customers to deliver personalised offers.

In the field of life sciences, GlaxoSmithKline’s data science and analytics teams use GPU-accelerated database by Kinetica for accelerated processing of transcriptions to run hundreds of thousands of chemical simulations across complex data sets for drug discovery and research and development.

What we foresee in 2018

We expect more financial services, retail and life sciences organisations as well as telecommunications, energy and logistics companies to leverage a high performance, in-memory distributed database to deliver truly real-time, actionable intelligence on large, complex streaming data sets.

Below are four technology predictions for insights-driven businesses to succeed in 2018:

1. Intelligent video analytics will play a pivotal role in the analytics market

2018 will be the year when we see intelligence videos carving out a larger share in the market. Ranging from surveillance cameras to smartphone cameras, most devices today are driven by cameras and smart sensors. Organisations will soon realise the importance of invaluable, real-time data collect from captured videos, be it by individuals or enterprises. For instance, studying some unique patterns of movement can indicate and help government and security companies reduce possible threats.

A retailer can account for accurate floor space, customer movement patterns, and adjust their shelf space based on customers’ interactions. It’s using real data to make decisions, instead of doing so on a hunch.

These pattern and geospatial insights garnered from video are almost readily available everywhere and any time – we will simply need to tap into them.

2. Asia Pacific will move from AI science experiments to operationalising it

In Korea and Japan, AI has already gone mainstream across industries such as banking and finance, telecommunications and retail. SK Telecom, for instance, announced in early 2017 that it will be investing US$4.2 billion in AI. In other APAC countries, including Singapore, Indonesia and Australia, AI is still in the early stage, but forward thinking organisations will adopt AI solutions that cannot be solved by traditional systems. The governments are pushing AI adoption – for instance, Singapore’s AI investment will exceed US$100 million from 2017 to 2022 in its bid to become both a smart nation and innovation hub.

As enterprises operationalise AI, they will look for products and tools to automate, manage and streamline the entire machine learning and deep learning lifecycle. Data scientists will need to focus on the code and algorithims and to deploy these they will require a enterprise class AI & BI GPU database that is operational ready for their company.

3. Modernising IT systems will meet the demands of the APAC population

The amount of data being generated will continue to grow in APAC – consumption is predicted to surge 30-60 percent per annum between 2015 and 2020. APAC will lead other regions in the amount of data being created largely from telcos, banking, e-commerce and social media. With 600 million people alone in ASEAN and a growing consumer base with wallet share, it creates a perfect storm of data for organisations to tap. However, the traditional data warehouse is increasingly struggling with managing how to service these consumers . Being on the forefront of growth also means being on the forefront of obstacles.

In 2018, organisations will start re-thinking their data warehouse approach to merge business intelligence (BI) and AI on a single platform. Such sophisticated platforms will not only help companies accelerate the processing of parallel data, but also derive highly accurate insights in real time.

4. Capture high-value customers with next-generation databases

AI is increasingly being deployed for various applications including personalised medicine and language processing. Enterprises will start investigating how they can accelerate the delivery of better services and develop more targeted campaigns to high-value customers.

Conglomerates, for instance, would be interested to know how to take care of their best clients and how to drive them to use their other services. Think a regular consumer of a supermarket, how can a conglomerate influence them to use their banking and health care services for their family? This is a market that requires deep understanding on real-time data to move consumers’ behavior. Responding to a consumer too late is a missed opportunity for them and the conglomerate.

2018 will be the breakout year in which GPU-powered databases–which are capable of processing data up to 100 times faster and only at 1/10 of the hardware costs required–are considered and piloted. Bringing AI and BI analytics together allows an APAC enterprise the ability to help their consumers get the best service outcome. It allows the enterprise to monetise and capture outcomes quickly.

Already, telecommunications, e-commerce, and financial services enterprises in China, Korea and Japan are deploying GPU-accelerated applications to counter data and analytics challenges. Financial services organisations can leverage the service for consumer and corporate banking. A algorithm output for credit score in a week to one delivered in sub-seconds changes an outcome for both consumers and the bank quickly. The GPU professing for AI has changed the IT and business landscape dramatically and its only starting to be recognised.