For more complete hierarchy of knowledge it is the data -> information -> knowledge -> wisdom. For more meaningful in the age of AI and data analytics point of view it is the descriptive - > diagnostic -> prediction -> prescriptive [1]. As you can see the information content referred to by Jane Austin's novels mentioned in the article is even at the lower part of the hierarchy. It is the job of the information theorists and practitioners to develop inference engines (e.g. GPT-3) that can provide the higher level of understanding and proper prescriptions. In addition even if the inference engines are not that capable yet, human-in-the-loop system can be deployed to significantly reduce the cognitive workloads of the human expert (e.g. cardiologist) to perform the better diagnostics for more accurate and timely prediction and prescription.
From data centric point of view in electronics and communication engineering Shannon's famous theorems on sampling and communication is a subset of data conditioning (to remove/overcome/mitigate noises on system) and data transmission (to remove/overcome/mitigate noises during transmission).
From TCP/IP networking point of view Shannon's theorem of communication, albeit a very important concept, is mainly at physical layer not even at the data link layer and the layers above (networking transport, application). The more universal information theory and entropy partly developed by Shannon is mainly belong to application layer including the AI and data analytics. For more complete treatment refer to the book by the late Dr. David J. C. MacKay entitled Information Theory, Inference and Learning Algorithms [2]. The author has made the book available from the website.
[1]https://www.ciscopress.com/store/iot-fundamentals-networking...