ibi Looks to Leverage Machine Learning to Improve Data Quality

ibi during a Virtual Summit 2020 event this week outlined a strategy that revolves around machine learning algorithms and other forms of artificial intelligence (AI) to embed real-time analytics into business processes.

Formerly known as Information Builders, the goal is to dramatically improve the trustworthiness of the data being relied on to prescriptively automate business processes in real time, says ibi CEO Frank Vella.

See also: The Strain on Data Supply Chains: Why Data Democratization is a Priority

At the core of that strategy is an Open Data Platform from ibi that enables applications to embed data visualization capabilities that are connected in real time to various data sources, file formats, and applications, including rival data visualizations tools from Microsoft and Tableau Software.

While ibi continues to provide its own Focus and WebFocus tools for visualizing data, Vella says rather than forcing organizations to abandon existing data visualization tools it makes more sense to invoke an enterprise-class data analytics backend engine developed by ibi via those tools. That approach enables organizations that have invested in data visualization tools that run on a local server to scale to the point where they can address enterprise-class analytics requirements, says Vella.

“Desktop visualization tools work for the individual, but they don’t work for the company,” says Vella. “They’re not enterprise relevant.”

Along with outlining its overall data management and analytics strategy, ibi this week launched Omni-HealthData Cloud Essentials, an analytics platform designed for mid-market healthcare organizations.

In addition, ibi announced a partnership with ASG through which data analytics and governance tools from both vendors will be integrated.

Vella contends organizations are not realizing the potential benefits of AI because the data on which the algorithms depend is unreliable. To address that issue ibi has been investing in machine learning algorithms that both identify data quality issues as well as surface recommendations for including additional relevant data within an analytics application.

The need for that capability has become more even pressing in the wake of the economic downturn brought on by the COVID-19 pandemic, adds Vella. Given the inherent uncertainty of the current business climate, Vella says the need for data that organizations can trust to make critical business decisions has never been more critical. However, that requirement has more often not resulted in organizations putting AI initiatives on hold because the data being employed to train AI models is simply inaccurate.

Vella says ibi has also been investing in a Project Unify initiative that will make all the connectors the company builds available via a multi-tenant cloud service. Via that effort, entire business processes and all their associated data will become a reusable microservice, says Vella.

Longer-term, those business processes and analytics are likely to be accessed both via speech interfaces and traditional graphical user interfaces (GUI), adds Vella.    

In the meantime, ibi is also committed to expanding the pool of talent familiar with its tools by partnering with academic institutions and providing free training to interns, adds Vella.

It may be a while before business executives trust the data collected in applications enough to automate processes at scale. However, each digital business transformation initiative presents an opportunity to address a data quality issue that has long plagued IT teams in general and business intelligence (BI) applications in particular. In the meantime, in the absence of fact-based analytics too many business decisions will continue to be based on intuition. The trouble with that approach is gut feel all too often winds up being a case of indigestion being mistaken for actual wisdom and insight.