Position Type:

  • Lassonde Undergraduate Research Award- summer research
  • NSERC USRA

Position Title:  Research Assistant/summer researcher

Location:  Petrie Science Building

Professor: Mojgan Jadidi

Department: Earth and Space Science and Engineering

Contact for Professor (Email, phone): mjadidi@yorku.ca or 416- 736 2100 x: 77704

# of positions available: 1 Student

Project Description:

Big Transit Data Analytical System

Big data analysis requires the integration of multiple sources of data (static and streaming), belonging to different types (geospatial, textual, temporal and numerical), and detection of any apparent patterns, concepts or features within the data sets. Monitoring the operation of a day in a transit system is one such big data case. With the rapid growth of computing power and intense increase of AVL (Automatic Vehicle Location) data collection and storage capability, Go Transit, like any other organization, has collected a vast amount of data about its business operation. This data needs to be mined to find useful information for business and service improvement. However, there are several significant challenges to turning large transit data sets into actionable knowledge, including the size of the dataset, the management of streaming data (such as GPS trace data), and the large number of logged people. Currently, GO Transit lacks a monitoring system capable of analyzing and aggregating the daily track of operations spatially, temporally and thematically.

The primary goal of this research project is to develop a big transit data analytical solution that will enable monitoring and tracking of daily operation through spatiotemporal aggregation solutions to support decision making under irregular conditions. OLAP (On-Line Analytical Processing) has been a popular tool for fast and user-friendly multi- dimensional analysis of data warehouses permitting to analyze data from different perspectives and with multiple granularities and aggregated levels. Indeed, Graph structures have been growing rapidly under umbrella of web semantics concepts in web-based application to analyze XML, social networks and spatiotemporal data; where not only individual entities but also the interacting relationships among them are important and interesting. In order to manage Transit operation data of GO Transit, an integrated approach of Graph database and OLAP system will be studied. We are going to explore the integrated graph-OLAP database concept through MongoDB, the leading modern database platform that has been a catalyst of the Big Data movement. This brings the advantage of adding the concept of Internet of Things (IoT) in order to access to stream sensor information coupling with transit data. In the same way, this pushes towards to Big data quantum leap, where transit operation data are coupled with relevant sensor information (e.g. environmental, human detector, crowd detectors, etc.). Enabling new type of analysis and mining solution on such huge amount of data helps organizations generate new revenue streams and achieve strategic goals. We need a motivated student to work on this big project that is started on March 2016. The proof of concept for visualization of such system was done in summer 2016. We need an enthusiastic student, who loves to play with data and aggregate them in the meaningful way to discover the change temporally, spatially or combination of both. 

Duties and Responsibilities of the student:

  • Literature review the current state of arts in Geospatial Big data integration and databases for Transit data and GPS Trace data
  • Work on big data aggregation algorithms
  • Develop a MogoDB-based database for such system
  • Connect the data to a WebGL visualization system which is developed already (Dynamic Visualization of Transit Data)

Skills and Qualifications:

(Technical Skills)

  • Understand the geospatial data structure
  • Familiar with geospatial databases (Mango DB)
  • Familiar with Sensor data structures
  • Web mapping   

(Interpersonal Skills)

  • Programing in Python, Java Script, HTML
  • ESRI product
  • Problem solving and algorithmic thinking

(Assets)

  • C or C++
  • Cesium or any WebGL based system
  • Analytical System design
  • Familiar with GPS trace data

Degrees, courses and Disciplines prerequisite*:  Any programing language courses

Stipend: TBD

Duration:  16 weeks minimum

Start Date:  05/01/2018 (estimated)

End Date:  08/31/2018 (estimated)

 

If you are interested in this research project, please contact Dr. Mojgan Jadidi at mjadidi@yorku.ca.