Could you please share your ideas about using deep learning for long-term forest management planning. Thank you all in advance.
I will answer in the affirmative. Indeed, for better planning of long-term forest management, universities must innovate through the dissemination of accumulated knowledge worldwide for long-term development.
I don't think that deep learning is, nowadays, the best tool for long-term planning. It needs a lot of data to be trained (not just a few hundreds or thousands of cases, but much more) and we don't have so many measured or even barely documented cases with pros and cons of different alternatives for long term forest planning.
There are other tools in machine learning (with less parameters, less layers, etc.) wich are less data-demanding, like tree decision or suport vector machine. Anyway, they must be trained (supervised) and they don't think or decide for you; they just "compare" new cases with train cases and provides you expectations for that new case based on what's happened in similar cases.
AI and machine learning walks towards not supervised techniques (I mean current papers; May and June 2020), that means 'not trained', but they are built on top of previously trained tools (although not specifically for the same task).
Hope this helps
Dear Professor,
Thank you so much for your contribution. I really appreciate it.
Regards.
Hi Dear Mehmet
Learning forest science should be continuous for all learners because nowadays we have large amounts of natural, man made disturbances.
and for all of them we should make decision as soon as possible.
Unfortunately foresters across the world used only production planning tools for forest management.The principles of natural planning was forgotten. Now being reversed after long drawn impact. So, first learning from nature, and then using local indigenous knowledge possessed by communities will make the syllabus of Long term forest management programme. carrying capacity based FM application will make the ecosystem stable.
- Badges
- Science topic