As an avid supporter of the chatbot development community, Sphere will feature a case study of a customer service chatbot they are developing in partnership with a Brazilian-based smart-loan provider,
Rebel, at the third international Chatbot Summit in Tel Aviv. This event will take place on January 31 at the InterContinental David Hotel in Tel Aviv, Israel.

In an effort to enhance the customer experience while simultaneously lowering costs, Rebel developed an AIML-based (Artificial Intelligence Markup Language) chatbot which drew information from a knowledge-base from interactions between customer service representatives and customers. Predetermined questions and answers allowed the initial chatbot to provide basic support. However, Rebel wanted their chatbot to be smarter, faster, and utilize an intuitive user interface before debuting the chatbot to its customers.

“When we first spoke with Rebel, they were using a basic XML interface for the chatbot,” said Sphere Software CEO Leon Ginsburg. “Rebel wanted an enhanced interface that would look great and be easy-to-use for both internal customer service representatives and their customers.”

The goal of the new interface was to successfully remove the “rough edges” of the chatbot by utilizing React.js and Angular.js technologies for the web-facing application, and Spring Boot and JavaScript for the back-end. A training function was also implemented into the interface so the chatbot could quickly decipher a “good” answer from a “bad” answer.

Rebel and Sphere worked together to further enrich the knowledge-base to be used by the chatbot. The chatbot was trained by utilizing the expertise of Rebel’s customer service representatives to provide quicker and more tailored answers. “Rebel had already established a fundamental domain specific knowledge-base for the chatbot,” said Ginsburg. “It was time to take that knowledge-base a step further so the chatbot could become smarter.”

Sphere partnered with Rebel to enhance their chatbot’s ability to learn and interact with customers through machine learning and an open NLP (Natural Language Processing) framework. For example, if a customer were to ask about the status of their loan, the Rebel chatbot would first ask for the loan ID and then open the back-end application to contextualize the loan ID.

The chatbot, which is still currently in development, will allow Rebel to achieve their business goals by reducing costs and creating a competitive advantage by adopting the technologies that will improve the speed and quality of their customer service. “This is a great partnership. We’re excited to implement a full-cycle chatbot development approach for Rebel through language parsing, syntax parsing, and machine learning technologies,” added Ginsburg.

Sphere Software will not only have a booth set up at the International Chatbot Summit, but will also be a proud sponsor at the event. Other sponsors of the third annual event include IBM, Oracle, and Vonage. For a list of keynote speakers at this year’s Chatbot Summit, visit their website
here.

You can learn more about how Sphere can help your organization implement chatbots by clicking here.