The Research And Application Of Data Mining For Road Transport Operation Management Under Big Data Background

Road Transport Operation Management under Big Data BackgroundIn recent years,with the prevalence of mobile Internet,Internet of things,computer technology and other emerging technology,information generated by industry applications present explosive development trend,the word ‘big data' suddenly becomes a new term for all walks of life.In the road transport industry,the information resources catalyzed by telematics,information technology and so on,showing the characteristics of scale,high speed,complexity and diversity.Through efficient development and use of road transport information's ‘processing' capacity to achieve its "value-added" function,to provide support decision,has become a tread.To this end,the article relays on the core information system in road transport,by data mining to improve industry operations management.As a multi-disciplinary fusion technology,DM is widely used in all walks of life.In this paper,through academic reference from domestic and foreign scholars on DM theory,methods and industry applications,through in-depth discussion for basic conditions of large data analysis,namely historical information,resource sharing,decision objectives,making sure the paper subject.Putting forward to apply cluster analysis,rough set and association rules to optimize the road transport management and operation,and get the effective verification.Based on passenger transport operation problems,that is,the transport mode fixing year around,DM is carried out from three aspects,that is,passenger route,departure date and moment.Through reasonable data preprocessing and indicators selection,taking Two Step method,constructing CF-tree prepares a dataset as a smaller subset,using hierarchical clustering to merge into the optimal categories number.Applying SPSS Modeler to build data flow model and achieve the visualization of results.Getting the regulation of passengers' travel route and time,and it provides decision-making basis for managers to carry out on-demand customization and flexible passenger operation pattern.Based on the transportation producers' assessment data in GPS,combined with rough set theory and association method,preprocessed the data to improve the rules quality,the algorithm is designed from data discretization,attribute reduction and rule building,which are achieved by information entropy theory,genetic algorithm and Apriori method.Through the reduction of conditional attributes to make sure the key affective factors,and reduce input information of rules.Finally,it helps transport operators and managers to strength the supervision of GPS by scientific and effective rules.In this paper,the main methods of data mining are applied to the analysis of information system data,which optimizes the macro-management and regulation of passenger and vehicle supervision in road transportation,and provides decision support for industry management.