Research On Recognition Of Urban Hot Scenic Spots And Recommended Routes Based On Optimized DBSCAN Algorithm

Taxis as an important means of transport in the city,its trajectory data contains massive city travel information,which is an important part of traffic big data.With the increase in the flow of people and travel demand in today’s society,discovering the knowledge contained in big data in the transportation field through data mining and other means and applying this knowledge to guide travel has become an important research in the field of transportation.Existing studies on travel planning for popular scenic spots generally focus on urban tourist attractions that are mostly located in the surrounding areas of cities,while there are few studies on scenic spots,hot-spots and recommended travel routes in urban areas.And the analysis based on traffic big data needs to be further improved.For example,the commonly used clustering technology cannot guarantee high efficiency and high accuracy at the same time when mining traffic hot-spots.At the same time,map navigation software usually only provides the shortest path between two points from the location of the tourist to a certain tourist destination,and cannot provide the optimal route containing multiple tourist attractions.Based on the above reasons,this article will optimize the DBSCAN clustering algorithm to efficiently and accurately dig out urban hot-spots from the traffic big data,and help people find interesting attractions more easily and quickly.Finally,the ant colony algorithm is used to realize the optimal travel route that can traverse the above-mentioned popular attractions.First of all,this article elaborates and analyzes several clustering techniques that have been applied in the field of traffic hot spot discovery.The shortcomings of the existing algorithms in practical application are put forward,and the corresponding strategies are obtained after comparative analysis,which provides a theoretical basis for the improvement and design of algorithms in the following article.At the same time,the quality evaluation criteria of the algorithm are summarized in order to evaluate the improved algorithm.Secondly,in order to solve the problem that the traditional DBSCAN clustering algorithm cannot take into account the efficiency and accuracy of mining traffic hot-spots at the same time,this paper combines the grid-based method and the density-based method.On the premise that only the grid size is required as a parameter,this paper grids the data set and performs density-based clustering on the grid data.After comparing experiments with other three algorithms,it is proved that the optimized algorithm has both the accuracy of density method clustering and the efficiency of grid method operation.And this algorithm further reduces the number of input parameters,avoiding clustering errors caused by too many parameter inputs like other algorithms.This feature is also one of the reasons why the algorithm is suitable for mining massive traffic data and discovering traffic hot-spots.Finally,this paper analyzes the trajectory data of urban taxis by clustering,and derives the traffic heat map of Chongqing City by time period,and also derives the ten most popular places in the urban area.By constructing the road network model of hot spots,this paper uses the ant colony algorithm to solve the traveling salesman problem to obtain the best and effective travel sequence and route.