life point, each iter if no relevant detected, life point -1
Start rule (a bigger problem when reusing model)
random sampling until ONE relevant study retrieved (Currently in use, can be unstable in extreme cases when seed set cannot represent the shape of the data)
Instead of random sampling, try stratified sampling from clusters. Add start rule of: at least one labeled example in each cluster.
Stop rule
Start rule (a bigger problem when reusing model)