Regulatory Operations in the era of AI

Artificial Intelligence is enabling businesses all over the world to create new value from their existing data.

Computer science defines AI research as the study of intelligent agents: any device that perceives its environment and takes actions that maximise its chance of successfully achieving its goals . More in detail, digital marketing experts Kaplan and Haenlein define AI as “ a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation ”.

For the regulatory areas in the life sciences industry, data mining – the practice of examining large databases to generate new information – is especially interesting because of the substantial amount of information that need to be controlled for quality and compliance purposes. This information is usually stored in many different formats and silos across the organisation, aspect that makes the process of collecting and reporting the information, as well as developing technology in support to it, a challenging and complex task.

Leading companies in this area have found that tagging documents with appropriate attributes (also known as metadata ) mined from the documents is one of the most compelling use cases. Typically, businesses have document management systems in place to manage documents across the organisation. Over the years, the document management systems have changed, the number of documents has grown exponentially and, upon examination, these companies often observe several of the following problems:

• Legacy documents migrated without accurate or complete metadata

• Documents assigned with incorrect metadata

• Changes in the vocabularies and terminologies leading to mixed metadata

• Changes in the document organisation, leading to documents with incorrect metadata

• Documents completely missing metadata (“unstructured” data)

• Changes in the organisation functions, leading to change in document categorisation.

Addressing these challenges can be overwhelming and feel insurmountable at times. That is why, nowadays, companies rely more on consultancies specialised in change management and facilitation of this process, which help businesses to:

• Define the desired metadata attributes (ie, the attributes for which the data should be mined from the document)

• Define the vocabulary related to the attributes required

• Provide methods to manage, correct and confirm the results (eg, migration process, validation, etc)

• Conduct data extraction and migration

• Achieve integration with other target systems.

The powerful capability of data mining means it should no longer be regarded as a niche solution; it is time for companies to explore the possibility of utilising its capabilities in standard processes. Data mining improves the quality and speed of every data-driven process and should be part of any company software solutions. At present it is difficult to assume the savings achieved, but users will see the benefit of cleared, well-connected and easily accessible information.

It is an opportunity worth exploring.