I have expertise in Image Processing, Machine Learning and Data Mining. I have difficulty in pinpointing a research problem. I have tried;
Feature Extraction and Feature Selection techniques for object classification.
Yield Forecasting from different indices calculated through remotely sensed images.
The problem with me is that I managed to publish some results but I was unable to select a problem of significant importance for my PhD. Any suggestion in this regard would be of help.
First you need to search some recent techniques and tools which are used by the researchers now a day. You may use ANN or Fuzzy logic with feature selection for some further improvements. You can also combine the big data platform with machine learning strategy. You may follow the below papers for your help,
http://www.sciencedirect.com/science/article/pii/S0031320316302655
http://ieeexplore.ieee.org/abstract/document/7437490/
http://pdgindia.com/uploads/accelerated_pso_swarm_search_feature27620162519.pdf
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7339682
Thanks, Philippe Fournier Viger, I have read the mentioned blog and learned some important insights for selection of a suitable topic. I am currently exploring some papers in NN regularization and activation functions and hopeful to come up with a suitable problem set.
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining