A Fast Algorithm for Association Rules Mining Based on Binary Search on Binary

To mine the frequent item sets from database conveniently and rapidly, a novel approach for association rules mining is proposed in this paper. In our approach, a vector subspace is build from database and the problem of searching frequent sets in database is transformed into that of searching vectors in vector subspace based binary search. Studies show that our approach is not only simple because it scans the database only once, but also has the virtues of reducing the size of vector subspace and accelerating the searching process.