Voting based taxi passenger-carrying point recommendation method

Abstract

The invention discloses a voting based taxi passenger-carrying point recommendation method, comprising the steps of: extracting historical passenger-carrying point data of all taxies, performing space-time clustering analysis to obtain a first candidate point set, performing time-space clustering analysis to obtain a second candidate point set, deleting all of the first candidate points and the second candidate points which are same in time attribute and have Euclidean distances smaller than a preset threshold value, generating a new candidate point through voting, grading and weighting, and obtaining a to-be-recommended candidate passenger-carrying point data set; and according to a target position and target time requested by a user, recommending candidate passenger-carrying points from the to-be-recommended candidate passenger-carrying point data set. The method can realize organic combination of space and time, avoids the problem of inaccurate passenger-carrying point analysis caused by difference due to time and space change, has the advantages of high time-space coupling degree and high accuracy, and can meet personalized demands of the user while recommending the candidate passenger-carrying points with relatively high passenger-carrying probability for the user.

Description

technical field [0001] The invention relates to a spatio-temporal cluster analysis data mining technology and a recommendation technology of a taxi loading point, in particular to a voting-based taxi loading point recommendation method. Background technique [0002] With the continuous development of GPS technology, mobile Internet technology, radio communication technology and other technologies, and the continuous maturity of location acquisition and mobile computing technology, it is becoming more and more convenient for people to obtain real-time location information of moving objects, and to collect space-time trajectory data of these moving objects The required cost is also getting lower and lower. For example, the trajectory of people's activities can be perceived through smart phones, and the trajectory of vehicles can be perceived through GPS sensors. The application of mobile positioning technology has penetrated into all aspects of people's lives, and the resulti...

AI Technical Summary

Problems solved by technology

[0008] The technical problem to be solved by the present invention is: aiming at the above-mentioned problems of the prior art, to provide a method that can realize the organic combination of spatial dimension information and time dimension information, and overcome the spatio-temporal correlation and spatio-temporal heterogeneity of spatio-temporal data leading to data mining accuracy Not high defects, effectively avoid the problem of inaccurate analysis of taxi passenger points due to differences in taxi passenger points due to time and space changes, good time-space coupling, and accuracy of time-space analysis results of taxi passenger points Voting-based taxi passenger point recommendation method with high speed and fast data mining speed

Images