Missing values are represented by question marks in Weka: ?

So, I think the easiest way to do is to change 99 by ? in your dataset

See the below example:

@relation 'cpu'

@attribute vendor { adviser, amdahl, apollo, basf, bti, burroughs}

@attribute MYCT real

@attribute MMIN real

@attribute MMAX real

@attribute CACH real

@attribute CHMIN real

@attribute CHMAX real

@attribute class real

@data

amdahl,29,8000,?,32,8,32,?

adviser,125,?,6000,256,16,128,199

adviser,29,8000,32000,?,8,32,253