The current resurgence of artificial intelligence (AI) technology may provide an antidote to the flood of data today’s digital world is facing because of the Internet of Things (IoT). Although still early, it seems that 2017 will be the year of chatbots and AI.
AI is changing the banking industry as we know it. Every day, banks are getting one step closer to supporting AI and conversational commerce, supporting projects that use chatbots to improve the overall customer experience. Last year Bank of America and Mastercard each announced the launch of chatbots (named Erica and KAI, respectively) which will allow customers handle routine transactions and financial advice, without any human interaction involved. Bank of New York Mellon was the first US bank to use bots in its everyday operations. The bank has programmed bots with rules that let them perform research on the orders, resolve discrepancies and clear the trades. It takes a human five to 10 minutes to reconcile a failure trade, whereas a bot can do it in a quarter of a second. Bots are also used in its data reconciliation group, doing all the work and only allowing humans to handle the exceptions.
The emerging use of AI and chatbots is going to shape the QA industry as well, with voice responsive test automation tools being the new trend. Machine Learning (ML), one of AI’s most common techniques, is all about enabling computers to recognize patterns in existing data and generate code that will make predictions on new data.
It is based on the usage of algorithms that learn from test assets to provide intelligent insights like application stability, failure patterns, defect prediction, etc, and answer questions like:
How can I measure QA performance? Will testing be completed on time? Is my project a healthy one?
It is based on the usage of algorithms that learn from test assets to provide intelligent insights like application stability, failure patterns, defect prediction, etc, and answer questions like:
How can I measure QA performance? Will testing be completed on time? Is my project a healthy one?
Conversational AI will become an integral quality assurance experience, and should work alongside human testers and not replace them.
More than 80 percent of testing is repetitive, often just checking that things work the same way as they did yesterday.
Working alongside AI and chatbots, testers will be able to better focus on the most interesting and valued aspects of software testing.
Bots will be in a position to understand complex questions, integrate with different data sources and take action for users to deliver the right outcomes on any device or interface.
More than 80 percent of testing is repetitive, often just checking that things work the same way as they did yesterday.
Working alongside AI and chatbots, testers will be able to better focus on the most interesting and valued aspects of software testing.
Bots will be in a position to understand complex questions, integrate with different data sources and take action for users to deliver the right outcomes on any device or interface.
The conversation can begin with the chatbot alerting the user that the test case execution has failed, then enquiring whether to retry or to check for wrong test data input. Then a machine-learning algorithm can autogenerate test assets can be much more efficient than a human tester. If an application has a hundred different conditions it is impossible to say that a manual tester can target every single one. This maximises test coverage and minimises redundancies and errors.
By using intelligent test data and feed with these the machine-learning software, the system will be able to identify which application components are at risk.
While robots and computers will probably never completely replace people, machine learning/deep learning and AI are transforming the industries, improving outcomes, and changing the way we think.
Tagged under