Mining Significant Utility Discriminative Patterns in Quantitative Databases

Drawing a discriminative pattern in quantitative datasets is often represented to return high utility (HUP). The traditional methods output patterns with above pre-given threshold. Nevertheless, the current user-centered algorithm requires outputting results timely manner strengthen interaction between mining system and users. Pattern sampling can probability guarantee short time, it could be candidate technology mine such patterns. In this paper, novel approach named HUPSampler proposed sample one potential HUP, which extracted significance according its database. introduces an interval constraint on length of HUP randomly extracts integer k proportion firstly; then, HUPs obtained efficiently from random tree by using growth way, finally, returns randomly. experimental study shows that efficient regard memory usage, runtime, distribution. addition, case studies show significantly used analyzing COVID-19 epidemic identifying critical locations.