Data infrastructure key to AI deployment for modern data landscape

Practicing scientists usually work with high volumes of to help organizations and institutions understand the bits and bytes of information they collect through various interactions with and stakeholders.

However, these professionals can only get started with their job when organizations have a clear understanding of their data, know how they collect it and where they store it, and are able to make it accessible to their AI and ML models for training and use.

In an interview with Tech News ahead of the ConnecTechAsia2019 Summit in in June, Axiata’s Chief Data Scientist Dr. Keeratpal Singh explains why must get the fundamentals right before diving into AI.

“When companies moved from paper to , they didn’t think about storage in the way they need to think about it if they have to implement an AI or ML solution.”

What Singh has put so simply is actually quite a massive challenge for many mid-market organizations embarking on their AI and ML journeys for the first time this year.

“In the modern data landscape, organizations need to hire data architects alongside data scientists because it is the data architect who will tame the massive data infrastructure that the company has built over the years and ready it for the sophisticated data models the data scientist creates.”

At face value, the challenge doesn’t seem too big because executives liken AI and ML models to yesterday’s intelligence (BI) platforms.

They tend to think that if the existing data infrastructure is suitable for BI platforms and can generate insights efficiently, it can be also be used to feed into AI and ML models. Unfortunately, that’s far from the truth.

“Previous BI systems only aggregated data and used blocks of historical data to create reports and dashboards. Today’s AI and ML technology needs access to the whole repository in order to spot trends and highlight unique data events that can be translated into a competitive edge for the business.”

The race is on for AI and ML

Some of the large enterprises such as banks, financial services organizations and government agencies are already using AI and ML and spending on the technology is soaring according to IDC and Gartner’s estimates.

Also Read:  The Key to Improving Global Cybersecurity is Data Science

“Early adopters in banking, retail, and manufacturing have successfully leveraged AI systems as part of their digital transformation strategies.

“These strategies have helped companies personalize their relationship with customers, thwart fraudulent losses, and keep factories running. Increasingly, we are seeing more local governments keeping people safe with cognitive/AI systems.

“There is no doubt that the predicted double-digit year-over-year growth will be driven by even more decision makers, across all industries, who do not want to be left behind,” said IDC Customer Insights & Analysis Research Manager Marianne Daquila.

That double-digit growth is primarily to come from mid-market enterprises who are just getting started with AI and ML and they’re the businesses that need to get the first steps right with their data infrastructure if they want to get the most out of their ROI.

“There are interesting things data scientists could do using AI and ML, but businesses need to be prepared to really leverage all their data to create that competitive edge and make the most of their data models,” explained Singh.

Lay the foundations for AI and ML now

Having been a data professional for more than half a decade, and with a strong foundation in engineering, Singh is of the opinion that the APAC region is well suited to leverage AI and ML in the coming months.

“We have the talent in this region and we have the financial resources. Further, our business leaders and regulators are both very forward thinking and keen on AI and ML.”

To be fair, AI and ML have been an integral part of the digital transformation agenda across industries since the beginning, and the success that some of the larger enterprises have achieved only serves as motivation for mid-market and even small businesses in this region.

Given the push on AI and the strong support from regulators across the region, Singh believes mid-market and small enterprises must quickly get to grips with their data infrastructure and prepare to make the most of their upcoming AI and ML deployments.

You might also like More from author

Comments are closed.