Case Study: Apply Ontology and AIGC for Task Appoint in Smart Building Control Centers

Since the amazing effect of ChatGPT has been recognized by the industry, towards integrating with employees’ workflow to create more business applications. At the same time, many artificial intelligence scientists are actively looking for methods and platforms that are highly efficient and can quickly embed LLM and gain business value quickly. This study proposes two important innovations. First, it uses historical operating data to quickly and automatically generate LLMs in the company’s exclusive domain. Second, to quickly embed those models in ontology maps that to generate thinking and decision-making documents (as AIGC). In the paper, first step is use auto-insight engine to transfer input data from operational datasets. Secondly, to adjust parameters the training in deep learning framework to output models. Finally, integrate more API that they like conversation tools or other services. It can automatically generate and output high-precision event determination results in real time and can automatic notification and complete the processing flow that integrate daily report for frontline partners. In future, it will be able to develop more follow-up applications, such as artificial intelligence assistant for the security control center partner, which can determine the high level of daily incidents and automatically complete high-complexity daily operation procedures in a closed-loop security environment, and moderately connect high-level voice assistants to achieve important indicators such as health care, efficiency improvement and cost reduction.