There are a few public datasets with textual security requirements data generated from IoT systems that you can use in academic research. Here are a few examples:

• The IoT Security Requirements Dataset (IoTSeRD) is a collection of textual security requirements for IoT systems, generated by security experts. The dataset is available on GitHub and is licensed under the Creative Commons Zero (CC0) license, which means that you can use it for any purpose, including academic research.
• The IoT Security Requirements Corpus (IoTSRC) is a dataset of textual security requirements for IoT systems, generated by both security experts and non-experts. The dataset is available on GitHub and is licensed under the Apache License 2.0, which means that you can use it for any purpose, including academic research.
• The Industrial IoT Security Requirements Dataset (IIoTSRD) is a collection of textual security requirements for industrial IoT systems. The dataset is available on GitHub and is licensed under the Creative Commons Zero (CC0) license.

In addition to these datasets, you may also be able to find public datasets of textual security requirements for IoT systems in specific domains, such as healthcare, transportation, and manufacturing. For example, the Healthcare IoT Security Requirements Dataset (HC-IoTSRD) is a collection of textual security requirements for healthcare IoT systems. The dataset is available on GitHub and is licensed under the Creative Commons Zero (CC0) license.

When choosing a public dataset to use in your research, it is important to consider the following factors:

• The quality and reliability of the dataset: Is the dataset well-curated and free of errors?
• The size and scope of the dataset: Does the dataset contain enough data to support your research? Does the dataset cover the types of IoT systems that you are interested in?
• The license of the dataset: Is the dataset licensed under a permissive license that allows you to use it for your research?