Picture: 123RF/alphaspirit
Picture: 123RF/alphaspirit

We’ve all heard “This call may be recorded for quality and training purposes” when contacting a call centre. And while some calls are used for training purposes, more often than not they are used to improve language-processing algorithms.

The insights gained from this information have driven businesses into a new domain of customer understanding, but limiting the analytics to this type of highly structured format excludes the majority of the data that’s being created.

Global research and advisory firm Gartner reckons 80% of the data created at present is unstructured, generated from conversations with customer service representatives and on social media sites, among other avenues. Now businesses are turning to natural language processing (NLP) technology to derive understanding from the countless unstructured data available online, in call logs and on social media.

Joining Michael Avery to discuss this evolving technology are Dr Vukosi Marivate, the Absa UP chair of data science at the University of Pretoria, and co-founder of Deep Learning Indaba; and Johan Steyn, Business Day contributor on tech and AI, and chair of the special interest group on AI and robotics with the Institute of Information Technology Professionals of SA.


Would you like to comment on this article or view other readers' comments?
Register (it’s quick and free) or sign in now.

Speech Bubbles

Please read our Comment Policy before commenting.