What is it about?
Modern smart buildings are packed with sensors tracking temperature, air quality, and energy use. However, the data they collect is often locked inside complex, disconnected computer systems that only technical experts know how to navigate. This paper introduces OntoSage, an intelligent chatbot designed specifically for buildings. Instead of digging through complicated software dashboards, building managers or everyday occupants can simply type a question in plain English—such as, "What was the average CO2 level in the conference room last week?" OntoSage automatically translates this question into computer code, retrieves the exact sensor data, runs the necessary math or analytics, and replies with a clear, easy-to-understand summary. By combining smart building blueprints with artificial intelligence, OntoSage makes it easy for anyone to understand, monitor, and improve their indoor environment.
Featured Image
Photo by Possessed Photography on Unsplash
Why is it important?
As buildings become increasingly automated to meet energy and health standards, the sheer volume of data they generate has become overwhelming. While consumer voice assistants (like Alexa or Google Home) are popular, they lack the deep understanding of commercial building infrastructure needed to answer complex analytical questions. This work is highly timely because it harnesses the rapid advancements in Large Language Models (LLMs) but solves their biggest flaw: "hallucinations" (making up incorrect information). OntoSage is unique because it anchors the AI to a strict, standardized digital map of the building (using Brick Schema), ensuring every answer is factually grounded in real sensor data. Furthermore, its lightweight adaptation workflow allows it to be deployed in completely new buildings with different databases without requiring massive retraining. This portability bridges a major gap in the industry, empowering facility managers and sustainability officers to make rapid, data-driven decisions without needing a team of data scientists.
Perspectives
Developing this framework highlighted a crucial realization: the true bottleneck in the smart building industry is no longer a lack of data, but a lack of accessibility. We frequently see highly advanced Building Management Systems that remain underutilized simply because their interfaces are too technical and intimidating for everyday users. My personal perspective is that natural language is the ultimate, universal interface. By forcing the AI to translate human intent into rigorous database queries—rather than relying on opaque, black-box predictions—we maintain trust, accuracy, and transparency. Ultimately, giving occupants and managers the ability to simply "talk" to their buildings democratizes environmental data, empowering everyone to actively participate in making our built environments healthier, more inclusive, and deeply sustainable.
Mr. Suhas Prakash Devmane
Cardiff University
Read the Original
This page is a summary of: OntoSage: Intelligent Human-Building Smartbot for Semantic Smart Building Question Answering, World Wide Web, January 2026, Springer Science + Business Media,
DOI: 10.1007/s11280-026-01403-0.
You can read the full text:
Contributors
The following have contributed to this page
