Objective: Effectiveness of data reduction directly influences the quality of reduced data,
design of reduction process and selection of relevant algorithms or model, thus ultimately affecting
flexibility and extension of the data reduction method and its implementation system. In view of the
deficiencies of the data reduction effectiveness evaluation system including imperfection and weak
applicability of index system, with lack of pertinence and neglecting the personalized needs of users,
three evaluation indexes were put forward to comprehensively reflect the rate of data amount reduction,
the rate of statistical difference and the rate of average information loss after data reduction.
Methods: On the basis of this, an evaluation method of data reduction effectiveness based on users’
interesting degrees was proposed.
Results: Through this method, subspace of the weight of indexes described users’ preference of index,
and it was accessed according to historical data.
Conclusion: The acceptable degree of reduction effectiveness is calculated approximately using the
Monte Carlo simulation, which implements the method of effectiveness evaluation of data reduction
geared towards different users and provides a quantitative basis for recommending data reduction
method for system users focusing on different aspects.