Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and system for forecasting shared bicycle traffic based on site behavior analysis

A technology for traffic forecasting and sharing bicycles, applied in forecasting, data processing applications, instruments, etc., can solve the problems of not being able to capture dynamic flow changes in time, and the accuracy of flow forecasting is not high, achieving high flow forecasting accuracy, reducing complexity, The effect of improving accuracy

Active Publication Date: 2021-07-13
SHANGHAI JIAOTONG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The invention cannot capture the dynamic traffic changes of the site in time, and the accuracy of traffic prediction is not high

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and system for forecasting shared bicycle traffic based on site behavior analysis
  • Method and system for forecasting shared bicycle traffic based on site behavior analysis
  • Method and system for forecasting shared bicycle traffic based on site behavior analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Such as Figure 1-2 As shown, the present invention provides a method for predicting shared bicycle traffic based on site behavior analysis. The site behavior is divided into inter-site circulation behavior and intra-site behavior changes, specifically including the following steps:

[0036] Data preprocessing steps:

[0037] Calculation of travel flow within a site based on historical itinerary information, including inflow check-ins and outflow check-outs, calculation of flow-ins and flow-outs between sites, and normalization; normalization of external factor data Processing to obtain real-time external input includes the following three steps:

[0038] (1a) Data cleaning: delete invalid data, where invalid data refers to data whose travel time is less than 1 minute;

[0039] (1b) Calculation of flow: calculate the inflow and outflow of each site, and the circulation data between sites.

[0040] (1c) Normalization: All site traffic is normalized by min-max, and the...

Embodiment 2

[0103] Such as Figure 1-2 As shown, the present invention provides a shared bicycle traffic forecasting system based on site behavior analysis, including a data preprocessing module, a model building module, a training model module, and a traffic forecasting module;

[0104] The data preprocessing module performs data cleaning, flow calculation and normalization processing;

[0105] The model building module includes inter-site flow modeling, dynamic intra-site behavior modeling and external influencing factor modeling;

[0106] Described training model module is to input training data into the built model in the model module, use Adam optimizer gradient descending algorithm to train model and obtain optimum parameter;

[0107] The traffic forecasting module inputs several historical site traffic data, inter-site traffic data and external factor data into the model in the training model module, and the model outputs the site traffic in the next time period, including inflow ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention provides a shared bicycle traffic forecasting method based on site behavior analysis in the field of traffic forecasting technology, including the following steps: S1, data preprocessing: including data cleaning, calculating traffic and normalization processing; S2, building a model: Including inter-site flow modeling, dynamic intra-site behavior modeling and external influencing factor modeling; S3, training model: input the training data into the model constructed in step S2, and use the Adam optimizer gradient descent algorithm to train the model to obtain the optimal parameters ; S4, traffic forecast: input several historical site traffic data, inter-site traffic data and external factor data into the model trained in step S3, and the model outputs the site traffic in the next time period, including inflow and outflow. The present invention models the dynamic circulation behavior between sites and the behavior change within the site at the same time to estimate the dynamic behavior change of the site, can capture the dynamic flow change of the site in time, and achieves higher flow prediction accuracy.

Description

technical field [0001] The technical field of traffic forecasting of the present invention, in particular, relates to a shared bicycle traffic forecasting method and system based on site behavior analysis, which improves the accuracy of site traffic forecasting by simultaneously modeling dynamic circulation behaviors between sites and behavior changes within sites. Background technique [0002] Traffic forecasting is of great significance to the construction of intelligent transportation systems, and can help governments plan ahead and travel companies to rationally adjust resources. In terms of shared bicycles, most of the existing methods use external factors such as weather, time and other characteristics to construct similarity functions or machine learning models such as random forests to predict future travel traffic at the site. However, these methods mostly use statistical methods when modeling the traffic behavior between stations, and they do not take into account ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06Q10/04G06Q10/06G06Q50/30
CPCG06Q10/04G06Q10/067G06Q50/40
Inventor 周纤沈艳艳黄林鹏
Owner SHANGHAI JIAOTONG UNIV