say or estimate that (a specified thing) will happen in the future or will be a consequence of something.
We have horoscope predictions, predictions of how a government may perform, predictions of what technology will bring, predictions that were made long ago in time about long ahead in future etc. There have been predictions like people would land on moon by Nostradamus which has been achieved. It was predicted that Iron plates would fly in sky. Business predictions, share markets, weather forecast, disaster management and the list go on and on.
It is about accounting a change with a probability. We also predict the success or failure of the prediction, learn from it and improve for the next set. There are parameters and they are weighted. They put together to give the final probability.
What has excited me more lately is taking the unpredictability into the account. How do we weigh the changeability while we predict? What amount of weight is given to unpredictability? This could lead to taking many new parameters into the account making the problem increasingly complex.
In a very natural and realistic way, taking unpredictability into the consideration does not make any sense at all. But however, when we see the bigger picture, not all problems have the same degree of unpredictability. That is where machine learning could help in deciding where to consider unpredictability and where not by mining the current web trends. Well, I am off to do more research on the issue!