I am working on a project to integrate data from Campus Management Systems and Learning Management Systems into a predictive AI model that forecasts students' academic performance. I want to use Microsoft Copilot for natural language processing and user query handling. What is the best approach to achieve this integration? Should I use open-source predictive AI models (like Scikit-learn, TensorFlow, or PyTorch) and then feed the results into Microsoft Copilot, or should I develop a custom Copilot in Copilot Studio to handle both predictive and generative tasks? Do you have any insights or recommendations on handling the integration?
Intelligent chatbots, which are in open access on the Internet, are so far not equipped with such advanced generative artificial intelligence technology, on the basis of which it would be possible to create highly complex, multi-faceted, multi-criteria, simulation-based predictive models with the help of which it would be possible to forecast complex processes. Such complex simulation models are built in other computerized information systems, such as Business Intelligence platforms equipped with analytical modules, Big Data Analytics, etc., and also increasingly in generative artificial intelligence technologies, but used for multi-criteria analysis of large sets of data and information.
I would like to invite you to join me in scientific cooperation on this issue,
Warm greetings
Dariusz Prokopowicz
Both approaches have their advantages. Developing a custom Copilot in Copilot Studio can deliver a more cohesive and streamlined user experience by harnessing the capabilities of Microsoft's ecosystem. On the other hand, using open-source predictive models offers greater flexibility and control over the modeling process. Ultimately, the best approach will depend on your specific needs, available resources, and long-term objectives.
- Badges
- Science topic