How do you generalise that the probability of "head" occurring on a coin is 0.5? Supposedly I toss a coin 1000 times a day for a period of 10 days. Each day the probability of head occurring on the coin changes. For instance - 670 times getting a head on day 1, 567 times getting a head on day 2 and so on. So each time, the probability of a head occuring isn't 0.5, so can someone please explain why we generalise that the probability of a head occuring on tossing of a coin is 0.5??
 A: Welcome to CV, Rhithik.
I think there is a foundational concept that you may need to understand a little better: when we say "something is random (like the value of a single coin flip, or of 1,000 coin flips)" in the statistical sense, what we mean is that the values of specific "realizations" (or "observations", etc.) cannot be known in advance—we can only speak of such events (single flips, or 1,000 flips) in terms of probabilities and probability distributions.
If we were to say "exactly 0.5, or 500 of these 1,000 flips will come up heads" we are no longer speaking of probabilities or probability distributions—precise articulations in the language of uncertainty—but of dead certainties.
So what does this precise value of 0.5 mean? In a probabilistic sense, it means we believe that a single coin flip has exactly as much chance of coming up heads, as it does of coming up tails. When we extend that belief about the behavior of a single coin flip to 1,000 coin flips, we use the binomial distribution to describe the probability of realizing 498 heads, or 499 heads, or 500, heads, etc. And if you play with the binomial distribution, you will find that while exactly 500 heads is the "most likely" realization for 1,000 flips, it is by no means the only possible realization (indeed, 498, 499, 501, 502 all have probabilities almost as high as 500). Still, that binomial distribution, parameterized by $p=0.5$ is our probabilistic description of $\boldsymbol{n}$ coin flips with probability of heads of $\boldsymbol{p=0.5}$ for an individual flip.
To sum up the "0.5" of the fair coin does not describe realized sample statistics (i.e. the actually realized number of 1,000 flips coming up heads on a given day), but rather describes a part of our conceptual model, part of the probability distribution of observing all possible values.
A: 
why we generalise that the probability of a head occuring on tossing of a coin is 0.5??

I don't think you can make that generalization unless you further assume that the probability of head on each day is itself a random variable with expectation 0.5 (e.g. one day is 0.6, another day is 0.4, etc. on average it is 0.5).
