Intelligent Recommendation Regarding Vehicles Suspected of Drug Transportation Based on FP-Growth Algorithm
The transportation of narcotic drugs serves as a crucial link in drug-related crimes.Despite the increasing diversity in transportation methods,highway remains the primary route for trafficking narcotic drugs,and different drug traffickers tend to employ distinctive patterns of drug transportation.This study conducts feature mining on drug transportation patterns,unveiling a correlation between the leading vehicle and the following vehicle.After analyzing and summarizing the characteristics of drug transportation on highways,it has been observed that in practice,there is a typical half-an-hour time interval between a leading vehicle and a following vehicle.Based on this finding,the modeling utilizes a PostgreSQL database,wherein three tables are designed for passing vehicles:the first half-hour intermediate table,the second half-hour intermediate table,and the intermediate span table.The technique of artificial intelligence data mining is applied to extracting the information on vehicle following patterns from a large volume of passing vehicles.Subsequently,the FP-Growth algorithm is utilized for mining frequent item sets with the objective of identifying frequently occurring license plates.By establishing thresholds and finding association rules,as well as filtering and sorting blacklists provided by anti-drug law enforcement officers,we can potentially provide recommendations regarding the suspicion level of drug transportation vehicles.This will support police's operation in intercepting vehicles suspected of drug trafficking,thereby enhancing the pertinence,accuracy,and effectiveness of suspect vehicles screenings to a certain extent.
transportation of narcotic drugspatterns of drug transportationfeature miningFP-Growth algorithmassociation rules
NETL
NSTL
万方数据
