Artificial Intelligence Vs. Human Intelligence
The ability to learn from experience or evolving intelligent thinking is the key characteristic of human intelligence. Now the question arises as to whether the machines can be made to carry out intelligent thinking without to human interface. Nowadays the global community is working using Artificial Intelligence to create machine man.
It is high time for people to know about this technology and its applications with its benefits to humankind and the implementation for effective planning and management of water resources available on the planet earth.
If we recognized that the key objective of AI technologies is to enable learning (from data), e.g. It is the key objectives of AI to solve problems, complete multiple tasks, and play with huge data, integration between human and machine, ability in planning and ultimate performance in water management. AI can be used to predict the risk of flooding beyond an acceptable socioeconomic threshold, forecast demand in a water distribution system, or estimate sediment transport rates in a river.
The key point here is that AI can be considered a way of creating useful models or methods to perform a complex task normally used to carry out by humans.
AI in Water Research and Practice
The global community has already benefited from the application of AI techniques in Hydro-environment research and practice. The Google search history shows an increasing trend with searching the keywords i.e. Machine Learning and Genetic Algorithm with “Water” and a huge number of research publications available on the internet being referred. More recently, a survey of ML methods for flood prediction indicated a trend of moving to ensemble methods and hybridized approaches where two or more ML techniques are used to predict the output variable. Widespread sensor deployment and availability of remote sensing data also offer new opportunities to hydro-environment practitioners.
They can help identify better model parameters, integrate ML with traditional mechanistic (physics-based) models or replace those when high speed of model execution is required. The use of deep learning methods in hydro-environmental practice is in a relatively early stage of development, however, the greater availability of data (and particularly big data through remote sensing) provides further opportunities for these types of AI methods.