Wise Practitioner – Predictive Analytics Interview Series: Kumaran Ponnambalam at Cisco – Machine Learning Times

By: Eric Siegel, Founder, Predictive Analytics World for Business

In anticipation of his upcoming conference presentation at Predictive Analytics World for Business Las Vegas, May 31-June 4, 2020, we asked Kumaran Ponnambalam, Analytics Architect at Cisco, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, Using Association Rules Mining for Segmentation and Profiling , and see what’s in store at the PAW Business conference in Las Vegas.

A Guide To Selecting Machine Learning Models in Python

Accounting for big dollars has Treasury embracing AI and machine learning

Seeq Expands Machine Learning Features for Process Engineering and Data Science Integration

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: My models deal with natural language understanding for contact center voice calls. They transcribe the calls and derive call summaries, intent and sentiment based on the transcriptions.

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

A: Predictive Analytics in our products help minimize and optimize the efforts of contact center agents. It provides them call summaries to minimize after call work. It also helps supervisors understand agent behavior and enable conformance to their procedures. It helps analysts to understand caller intent and sentiment during and after the call.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: Our call summaries help minimize after call work for agents. It is expected to reduce after call work times by at least 50%.

Q: What surprising discovery or insight have you unearthed in your data?

A: The traditional assumption in contact centers is that all agents are equal in skills. Hence, calls are usually distributed to agents in round robin fashion. We found that the skills of agents vary and they do influence the outcomes and sentiments of calls.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: Association Rules Mining is traditionally applied for classification variables only. By doing a few adjustments, they can be used to analyze continuous data and more specifically find deviations in metrics across time.

—————————–

Don’t miss Kumaran’s presentation, Using Association Rules Mining for Segmentation and Profiling, at PAW Business on Wednesday, June 3, 2020 from 4:20 to 4:40 PM. Click here to register for attendance.

By: Eric Siegel, Founder, Predictive Analytics World for Business

Credit: Google News