Subway passenger-oriented friend recommendation method and system

Abstract

The invention relates to a subway passenger-oriented friend recommendation method. The method comprises the steps of obtaining source data of subway station passengers and subway operation; preprocessing the acquired source data; inferring a passenger travel path according to the preprocessed source data; calculating a detailed train operation time table according to the subway train time table and the train departure interval; matching the passenger with the specific train according to the inferred passenger travel path and the calculated train operation time table; and extracting and measuring the interaction condition between the passengers according to the matched passengers and the specific train, and carrying out friend recommendation. The invention also relates to a subway passenger-oriented friend recommendation system. According to the invention, original social software friend recommendation based on user attributes or static space is expanded into dynamic space, so that matching between passengers and trains is more accurate, and interaction conditions of the passengers in the subway system can be comprehensively quantified.

Description

technical field [0001] The invention relates to a friend recommendation method and system for subway passengers. Background technique [0002] As a closed space where a large number of people gather, urban public transportation systems (including bus systems and subway systems) are prone to interactions among passengers, and there are a large number of "familiar strangers" among passengers. Studying the interaction within the public transportation system is helpful to discover passengers with common travel characteristics, dig out the commonalities of social attributes and behavior preferences, and improve the recommendation effect of mobile social software. [0003] The passenger interaction research in the existing urban public transportation system is mainly concentrated in the bus system, and the subway system has not been fully studied. Compared with the public transportation system, the subway system is larger and more complex. There are many challenges in extracting ...

AI Technical Summary

Problems solved by technology

However, the subway card data only has relevant records of passengers entering and leaving the station, and lacks the description of passenger behavior within the subway system, such as where to transfer, which train to take, etc., which are very important for measuring the interaction between passengers and making friend recommendations is very necessary

The existing matching algorithm for passengers and trains is to directly estimate the corresponding trains they take based on the passenger’s time of entering and exiting the station. The results have certain deviations and are not accurate enough.

Images