Transforming Student Engagement: CloudMile and NTU Collaborate to Enhance AI-Powered Virtual Assistants
Siri, Alexa, Google Assistant: whatever virtual assistant you use, we can agree that they are now a commonplace part of modern life.
It is staggering to think about how quickly artificial intelligence has grown in the last decade, making its mark across multiple industries. There was a time not too long ago when the humble chatbot was still a novelty.
Virtual Assistants: From Basic Bots to Advanced AI
In the 80s and 90s, chatbots were in the early days of development, mainly used for research purposes and answering simple questions. These early text-based bots worked by following strict rules and scripts to respond to users. As the Internet grew in the 2000s, chatbots became more popular for helping businesses with online customer service, offering quick responses to common queries.
By the mid-2000s, Cleverbot emerged, bringing a more conversational approach that allowed for fun, free-flowing interactions. The 2010s marked a turning point, with advances in AI and machine learning enabling chatbots to handle complex conversations.
Fast forward to today, Large Language Models (LLMs), like Google's Gemini, are enhancing these virtual assistants even further. LLMs can understand and generate human language, making them powerful tools for writing, customer support, and more. They help businesses automate tasks, personalize interactions, and improve productivity by acting like smart assistants that enhance communication and information access.
NTU’s Innovative Chatbot Journey
In 2020, Nanyang Technological University, Singapore (NTU Singapore) launched a virtual assistant chatbot to onboard 6,000 freshmen, which also helped reduce in-person resources while maintaining good interactions with freshmen during that time. However, the Lyon chatbot, built on Google Cloud’s Dialogflow Essentials, was limited by its intent-based design and struggled to keep up with changing user demands.
The university faced challenges in managing the vast amounts of data required to train the chatbot effectively. Data training, a meticulous process of correcting inaccuracies and inconsistencies, often burdened teams with ensuring the chatbot operated on accurate information.
To address these issues, NTU decided to upgrade their chatbot with expert consultation and tailored solutions from CloudMile in Singapore. Ng Kee Haur, Deputy Director at NTU's Centre for IT Services, explained, “We are transitioning from an intent-based Natural Language Processing chatbot to a GenAI chatbot for student housing inquiries to enhance flexibility, scalability, and user experience.” He added, “While GenAI can handle complex, unstructured queries, there is still a role for the intent-based chatbot in specific scenarios. This hybrid approach allows us to balance efficiency with precision.”
CloudMile and NTU: Pioneering a New Era of Chatbot Efficiency
As Generative AI (GenAI) gained momentum, CloudMile helped NTU enhance their chatbot, making it faster and more efficient at handling student housing queries. Together, they designed a smarter conversation flow that breaks down information collection and processing into key steps: understanding the query, rewriting it, classifying user needs, and generating clear answers.
By leveraging Google’s Gemini LLM, the chatbot now pulls information from various centralized, curated data sources, ensuring users receive accurate and up-to-date responses. This integration provides robust checks and balances, resulting in reliable, contextually accurate answers while minimizing errors.
These enhancements contributed to high chatbot accuracy and overall effectiveness, leading to a smoother experience for both students and staff. “With the new AI-powered chatbot, we are currently saving around 14.5 days’ worth of work every month,” said Alvin Ong, Chief Information Officer of NTU. “This allows our staff to focus on tasks that require human insight and creativity.”
Jeremy Heng, SEA Managing Director of CloudMile, remarked, “The integration of GenAI technology into NTU's Lyon Bot, marks an exciting step towards improving student to lecturer engagement. This collaboration underscores our commitment to using advanced technology to optimize processes and elevate the quality of human interaction across the educational institutions.”
Want to learn how to drive your customers’ AI transformation and achieve business success together with CloudMile? Click here.
