The main challenges include are:
1. Model performance during extreme conditions
2. Non-randomized nature of data
3. Dealing with many inputs of ground data
Topic is: Time series crop yield estimation using satellite images
If my PS is wrong please also help me to correct it
Dear Hamza Mehmood
Just focusing on literature review and what your contribution in your topic of thesis "Time series crop yield estimation using satellite images"
Kind Regards
Qamar Ul Islam
In general, the defence of the problem statement can be approached along the following guidelines:
1- To identify a gap in the literature review and tell your audience how your are going to fill this gap by your research.
2- Propose the use of a different approach to tackle a known problem and this is based on the literature review and you need to articulate the expected benefits of this new approach.
3- Also you can carry a comparative study of existing approaches using similar or special data set.
Please check my Researchgate for a document titled how to develop and write your proposal and you will find some highlights about the problem statement.
Regards,
Gamal
I suggest to go one step back and reframe your problem statement within a well-defined research methodology that you fully understand. Doing so would clarify all the elements that support the need for your proposed solution.
Can the problem have multiple (valid) solutions? (i.e., your solution to the problem is not the only way to solve it). If yes, then go for the Design Science Research Methodology. If your topic has to do with software, computer science, or data science, then this book can be extremely helpful:
- Wieringa, R. J. (2014). Design Science Methodology for Information Systems and Software Engineering. Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-662-43839-8
Within this methodology, you formulate your research question with 4 elements. The above-mentioned book suggest the following template for the definition of a research problem:
• Improve
• by
• that satisfies
• in order to .
The research question becomes (in general terms) how to solve this research problem?
For a quick explanation of the methodology (before jumping into the book), the author also provides a series of videos on youtube
- https://www.youtube.com/watch?v=ipPam1k_KNg
Sometimes the term "artifact" is too generic and could be misleading. I find this classification of possible artifacts useful:
- Offermann, P., Blom, S., Schönherr, M., & Bub, U. (2010). Artifact Types in Information Systems Design Science – A Literature Review. In R. Winter, J. L. Zhao, & S. Aier (Eds.), Global Perspectives on Design Science Research (Vol. 6105, pp. 77–92). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-13335-0_6
In case the solution to your problem is unique, then you are dealing with empirical research problem, and another methodology with focus on such questions should be followed.
- Badges
- Science topic