ML and AI: changing how data science is leveraged for digital transformation

The growth and success of companies including Amazon, Facebook, Google and Uber which operate on digital platforms raise fundamental challenges for executives of established companies.

The digital natives are surfing a wave of massive change, dubbed by some as the Fourth Industrial Revolution, and recognised by Klaus Schwab, Executive Chairman of the World Economic Forum, as change that “is disrupting almost every industry in every country”.

For established companies to thrive, executives must formulate strategies that drive digital transformation, so their customers’ experiences and their costs of operation match those of the disrupters.

>See also: What is machine learning?

Companies that become AI-driven will greatly increase their chance to succeed in these times of change and disruption. Organisations use artificial intelligence, specifically machine learning, to predict what is most likely to happen next based on previous experiences, and then act to drive outcomes that satisfy customers and grow revenues.

Data is digital transformation’s primary resource. Sufficient investment in infrastructure is necessary to ensure that this resource is first harvested, and then fully utilised.

However, there are three major challenges. The first is the shortage of data scientists — the experts with programming skills, statistics and machine learning knowledge, and an understanding of the business.

Finding these experts is proving challenging and this shortage is stifling machine learning initiatives. The second is time to value: machine learning projects can take weeks or months to deliver the predictions needed for digital transformation. The third is ensuring the quality of the resulting model — vital to guaranteeing the actions taken impact earnings.

DataRobot, the pioneer of machine learning automation, was founded in 2012 to address these challenges. The company hired some of the leading data scientists on the planet – Kaggle competition winners – and set about teaching the machines how to do the job of a data scientist.

>See also: AI: the greatest threat in human history?

In doing so, DataRobot also opened the doors to business executives allowing them to use machine learning predictive analytics directly without involving data scientists. But the path to becoming AI-driven involves more than automation.

Success entirely depends on selecting the right projects; and so DataRobot University was launched to teach business people how to select good projects that will have real business impact.

Automation is revolutionising data science. Executives who cannot hire their way out of the data science shortage now use a platform that automates the process of deploying machine learning applications.

This allows established companies to use their strengths – the data they have about their customers and their people experienced in operating their business processes – to put digital transformation into the hands of these experienced staff.

At the same time innovation in open source analytic languages, such as Python and R, is moving fast. Here, machine learning automation plays a big role in de-risking open source adoption. This is achieved by providing enterprise readiness features that users are accustomed to in expensive proprietary systems, such as SAS and IBM SPSS.

>See also: Data scientists are the key to solving real world problems

For example, extensive information about algorithm behaviour, enterprise model deployment and refresh, security and model sharing. This simplifies integration with Big Data stacks and reduces cost of ownership.

Digital transformation in established companies is a team effort involving a diverse community of people with different skills and expertise. This includes data scientists, business analysts and business users.

While expert data scientists have historically been forced to write custom code, automating this task empowers them to shift focus from routine coding and data wrangling to tasks that add real value: understanding the business problem, the deployment context of their models and explaining their results.

Business analysts with deep statistical training but limited programming skills need guardrails, so the software must actively guide them through the process of developing and deploying their machine learning models. Business users want to engage in the model development process and then visualise the characteristics of a model to understand how it behaves, so they can explain decisions to their customers and managers.

Machine learning automation software must also scale to the enterprise level as measured on many dimensions, including by numbers of users, by projects, by models, and by data volume.

In practical terms, this means that the software should support deployment in Hadoop, provide standards-based integration with databases, and allow low maintenance provisioning, either on premises or in the cloud.

>See also: Digital transformation: an analysis of the potential and the challenges

Applying automation to data science is a transformative process, and is best undertaken in stages. Start by identifying manageable yet impactful initial projects, develop focused small teams, move and learn quickly, celebrate success, and then scale-out as organisational skills and business needs dictate.

Ultimately, the success of organisations transforming to AI-driven enterprises lies in putting data science in the hands of business users. DataRobot is democratising data science by teaching the machines to do the work of the data scientists so business users take direct responsibility for deploying sophisticated machine learning algorithms into the business processes they operate.

Automation creates powerful and profound impacts on business and on data science. Embracing this new technology gives companies a valuable tool in their quest for digital transformation.

Now that the automated machine learning genie is ‘out of the bottle’, time is of the essence. Established companies that move fastest to become AI-driven will take full advantage of their strengths as they defend against digital disrupters.


Sourced by Yvonne Cook, European sales director, DataRobot


The UK’s largest conference for tech leadership, TechLeaders Summit, returns on 14 September with 40+ top execs signed up to speak about the challenges and opportunities surrounding the most disruptive innovations facing the enterprise today. Secure your place at this prestigious summit by registering here

Avatar photo

Nick Ismail

Nick Ismail is a former editor for Information Age (from 2018 to 2022) before moving on to become Global Head of Brand Journalism at HCLTech. He has a particular interest in smart technologies, AI and...