Domain is the set of all values that can be input into the function to get a meaningful result, or in the case of a graph, the set of all possible x values. The range is the set of all outputs for the function, which on a graph is the set of all possible y values.
For the function f(x)=x, also known as y=x, both the domain and range include all values between positive and negative infinity.
For f(x)=1/x, the domain is everything but zero, and the range is also everything but zero.
For f(x)=x^2, the domain is all values between positive and negative infinity, and the range is all values greater than or equal to zero.
To find domain, investigate whether any inputs will give nonsensical or undefined results, and exclude those. What remains is the domain.
To find range, investigate whether there are any values that will never be output from the function, and exclude those. What remains is the range.
For intercepts, you look for where in the function either x or y will be zero. When x=0, that's the y intercept. When y=0, that's an x intercept. Note that a function can potentially have any number of x intercepts but by definition can only have one y intercept. If you're not dealing with a function, then this restriction is removed.
Hope that helps. Check out Wikipedia and Wolfram Alpha if you want further references. Good luck!