A Parallel FP-Growth Mining Algorithm with Load Balancing Constraints for Traffic Crash Data

Traffic safety is an important part of the roadway in sustainable development. Freeway traffic crashes typically cause serious casualties and property losses, being a threat to public safety. Figuring out potential correlation between various risk factors revealing their coupling mechanisms are effective ways explore identity freeway crash causes. However, existing association rule mining algorithms still have some limitations both efficiency accuracy. Based on this consideration, using data obtained from WDOT (Washington Department Transportation), research constructed multi-dimensional multilevel system for analysis. Considering load balancing, FP-Growth (Frequent Pattern- Growth) algorithm was optimized parallelly based Hadoop platform, achieve efficient accurate calculation massive amounts data; then, according results mechanism among precursors, causes were identified revealed. The show that parallel FPgrowth with balancing constraints has better operating speed than conventional FP-growth towards processing big data. This improved makes full use cluster resources more suitable large sets while retaining original advantages algorithm. In addition, rules model improvement interaction proposed can catch occurrence consequences (lower support degree probably) accurately efficiently.