In Java please

bayesianAverageRating $-$ in the ratings. Ratable, add a method named bayesianAverageRating that takes $2$ ints

and returns a double:

The first parameter is a number of extra ratings

The second parameter is the value of the extra ratings

The method returns the average rating of the song$/$movie if the extra ratings were added. Note that you are not

actually adding the ratings to the Song$/$Movie$.$ You are only using the extra ratings in your computation of the

average

Example: If a song has ratings of $4$ and $5$ and this method is called with parameters $2$ and $3(2$ extra ratings of

value $3),$ then the bayesian average is $\frac{4+5+3+3}{4}==3\mathrm{.}75$ instead of the regular average of $4\mathrm{.}5$

If a song$/$movie has no ratings and this method is called with $0$ additional ratings, return $0\mathrm{.}0$

The method should work as expected in the edge cases of $0$ extra ratings $($eg$.$ Adding $0$ extra ratings of value $3-$

the output should be the normal average rating$),$ and a song$/$movie that has no ratings $($eg$.$ Adding any number

of extra ratings of value $x$ to a song$/$movie that has not been rated should return $x)$

If the input is invalid, this method will return $0\mathrm{.}0.$ The input is invalid if the value of the extra ratings is anything

other than $1,2,3,4,$ or $5$ OR the number of extra ratings is negative

This method will be inherited by both Song and Movie, though you should test this method through the Song

and Movie classes. It is acceptable, and expected, that you copy your tests and use the same exact test cases for

both Song and Movie when testing this method.

Why? When comparing songs $($Or anything$),$ it's common to have many songs that are only rated by a single

reviewer who gave it a $\frac{5}{5}.$ This song would have a perfect average rating of $5\mathrm{.}0$ and would be rated higher than

a song that has been rated $100$ times with an average rating of $4\mathrm{.}99.$ The second song is well$-$loved by many

people, and you would $[$likely$]$ rather have that song in your playlist than the song that only $1$ person likes and

no one else is aware of$.$ We adjust this by adding a few fake ratings to each song. In the example of $2$ ratings of $3$

being added, the first song would be lowered to an average of $3\mathrm{.}67$ while the second song's rating would still be

$4\mathrm{.}99.$ This provides a more useful rating system than directly taking the average of all ratings.