spamassassin,postfix and Mailscanner

I don't think anyone can tell you if these values are correct - you should decide by yourself if they are ok for you.
I would look on the spam you are getting, and try to tune some other rules for spamassassin, or even add some new ones.
I don't use spamassassin system-wide - I just use it on my workstation. I added new rules, that fit my spam emails. This can be painful job to look at all your spams, but I don't think you will have to do it frequently.
Ie. if your spams contain word pharmacy, rule to give some extra points for this would look:
body BODY_PHARMACY /pharmacy/i
describe BODY_PHARMACY pharmacy word found
score BODY_PHARMACY 0.4

Just look at your spams and write new rules or tune a little (to give some more points) existing ones.

+15 points for 99% probability is a lot I think (as your required score is 3.5). Moreover - giving -15 for 0% can cause that spams are getting through your barrier. I didn't change bayes scoring below 50% - they are good for me. I change just 50% and more to:
score BAYES_50 1.2
score BAYES_80 3
score BAYES_95 4.2
score BAYES_99 5
My required score is set to 5.
But - these values are good for me (and I have more rules changed from their default settings), YMMV.

Thank you very much for the reply , please tell me where to add the following
body BODY_PHARMACY /pharmacy/i
describe BODY_PHARMACY pharmacy word found
score BODY_PHARMACY 0.4

thankyou.

The same file, where you added your score definitions for BAYES_00, and so on.

Thank you ,
if i add the following:
body BODY_PHARMACY /pharmacy/i
describe BODY_PHARMACY pharmacy word found
score BODY_PHARMACY 0.4

then the mails which contains the text  pharmacy will be considered as spam , please suggest me.

From now on (after adding this rule), all emails containing 'pharmacy' will get extra points (0.4 in above example). You can add more rules with other words or phrases and assign them some points.
Detailed instructions on creating your own rules are described in documentation. Just run this command, to read it:
perldoc Mail::SpamAssassin::Conf

Thank you very much , i will read that , for now i have added the following in the /etc/mail/spamassassin/local.cf
score BAYES_00 -15.0
 score BAYES_05 -5.0
 score BAYES_95 5.0
 score BAYES_99 15.0
will this be okay for my setup , please suggest

I just wrote it in my first comment - it's hard to say if these rules are good for you. The only person who can know this is you :)
I can only suggest to increase points (make it closer to 0) for BAYES_00 and BAYES_05 (or even delete these two rules).

Sir,
as said in my previous post
#  Set the threshold at which a message is considered spam (default: 5.0)
required_score 3.5
and after adding your score
score BAYES_50 1.2
score BAYES_80 3
score BAYES_95 4.2
score BAYES_99 5
now those spams are not getting trrough the users mailbox instead the same is moved to a archive mails which i had set in the mailscanner
hence i have decided to takeup your values  will there be any problems if the following values are kept
required_score 3.5
score BAYES_50 1.2
score BAYES_80 3
score BAYES_95 4.2
score BAYES_99 5

Iam sorry for asking this as iam new but the service is required to be good untill i read this completely hence please suggest me will the above settings work

Will they work? It depends on what do you mean saying 'will they work'.
Yes - if your bayesian filter will say that there is 99% chances that email is spam, then spamassassin will give such email 5 points extra. If this email won't get any negative score, it will have 5 points. As you have required_score set to 3.5, such email will be considered as spam.
I don't know what are you doing with emails which are recognized as spam. If you are rejecting/deleting them, then there can be a problem, when some good email is tagged - but it can always happen - there is no ideal filters, which do not make mistakes.
If you don't reject/delete spam, just tag it or move it to some spam/junk folder, there is nothing at all you can worry about - the worst case is, that good mail, recognized as spam, will be in your spam/junk folder.

You have to remember, that there is always possibility, than some good mail will be considered as spam. These are so called 'false positives'.

thank you very much , the emails whiich is recognized as a spam are getting to a mail accout where i keep checking .
Sir, can you please provide me a docs or weblink so that i can read and understand completely about the BAYES SCORE settings and it is usage.
thanks.