Avik Chatterji

Topic: Non-Functional Testing of Chatbots

Abstract: Digital has ushered in an era of technology trends such as the Internet of Things (IoT), Artificial Intelligence (AI), intelligent apps, cognitive computing, deep learning, machine learning etc. Businesses are leveraging these to succeed and maintain a competitive edge in digital. As the focus turns to driving more personalized customer experience, non-functional testing (NFT) becomes crucial, and NFT engineers must gear themselves for the technology shift.

Of these technologies, the “talking” Virtual Personal Attendant (VPA)/ Chatbots is gaining traction among businesses. The new generation of this autonomous technology aspires to give the computer interface a more “humane” approach, something similar to an interaction with an actual human, who can provide intelligent solutions, answer questions and solve everyday problems with a personalized touch.

Chatbots combine machine-learning algorithms with automation software, which helps them understand, evaluate and process the queries to provide intelligent solutions with great speed and impeccable accuracy.

VPAs are often exposed to all-important PII’s (Personally Identifiable Information), financial details of the users, making them susceptible to theft and impersonation. Security, thus, becomes an important aspect of a successful chatbot rollout. Additionally, performance also differentiates one Chatbot from another.

This paper talks about VPAs, their types and working methodology. It goes on to further explain the need of Performance & Security Testing for AI-based apps like Chatbots, its typical use cases, the testing strategy and tools for testing these kind of applications. This paper will further go on to elaborate how we can embrace Artificial Intelligence and VPAs to develop leading edge Non-Functional Testing services for our clients apart from traditional Non-Functional Testing.