CHATGPT APPLICATION IN POWER BI: A CASE STUDY
Keywords:
Natural language processing, Transformers language models, Power BI, Business intelligenceAbstract
This article presents the results of a study that evaluated the impact of integrating the ChatGPT conversational artificial intelligence model into business intelligence reports generated with Power BI. The goal was to determine if ChatGPT could improve the accuracy, clarity, and usefulness of the narratives and ideas included in these reports. From the theoretical framework, the potential of generative AI to support human capabilities in data analysis is highlighted. The methodology consisted of a pre-post quasi-experimental design with group control applied to a sample of 30 reports from a company. The results indicate that integrating ChatGPT using Power Automate contributed to significantly improving the accuracy and clarity of findings compared to regular manual reporting. Although there will not be a statistically significant improvement in the number of actionable ideas generated, the potential to train the model with business knowledge will be appreciated. In conclusion, initial evidence suggests that adopting conversational AI solutions like ChatGPT can bring value to enrich analytical capabilities in the business intelligence reporting process. It is recommended to complement it with specialized logic in the business context.