As I want my research to accurately represent Big Data context, I am thinking of avoiding only the use of big datasets, as it has been the case in many Big Data related research. My aim is to represent the 3 v's of Big Data correctly. How should I proceed? Thank you for sharing your opinion.
"Avoiding the use of big datasets", I think you are on a correct track. I suggest the "velocity" perspective of the big data. If you have computing background, I can suggest to developing data pipeline to collect and process the "run time" data taken from patients/devices. Especially from microservices and serverless development perspective, you can propose a fast run-time system for your research. Sorry if that sounds too technical for you.
Thank you Mehmet. Your answer makes sense. One of the option I am investigating is to setup cloud based real time data lakes.
To aim at volume and variety (with a semantic accuracy) one option would be integrating different distributed Linked Data sources. Persistence may be supported via triple-stores or other (NoSQL) DBs based on different data models, that often offer standardized endpoints.
One of the many areas in which the Big Data research or better still search for insightful and accurate analysis of captured datasets have been applied is within the field of Process Mining. The process mining has proved itself to be the missing link between the model-based process analysis and data-oriented analysis techniques. Perhaps, looking at the healthcare context which your research considers, I will suggest you employ methods that supports the semantic-based process mining. The semantic process mining brings the idea of using semantic technologies in combination with the process mining techniques to manage the perspectives of any given process (e.g. the healthcare domain).
- Badges
- Science topic
- Similar topics
- Cognitive Science
- Thinking