Spatial-data-driven student characterization: Trajectory sequence alignment based on student smart card transactions

Sungha Ju, Sangyoon Park, Hyoungjoon Lim, Sung Bum Yun, Joon Heo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Analyzing students' characteristic can provide much information for campus planning, education design and student management. This study built students' sequential trajectories based on student smart card transactions and calculate similarity scores for finding relationship between students' trajectories and academic performance. The data used in this study are student smart card transaction data and attendance information of Yonsei university Songdo campus students. Based on this, the trajectory of each student is created into daily context sequence and connected in semester unit. In order to calculate the similarity of one semester trajectory between two students, Needleman-Wunsch Algorithm, which is mainly used for comparison of the DNA nucleotide sequences of two different species, was applied. The similarity score of trajectory sequences for student pair were calculated for 685 students in spring semester. For finding relation with academic performance, authors divided students into two groups; one group with high similarity score for both students in the pair and the other with pair of students with low similarity score. 2-sample T-test was conducted afterward in to determine whether the GPA of these groups were different form the overall distribution of student GPA. As a result, the mean value of GPA of the students with low similarity scores were statistically significantly lower than the overall mean value of GPA. This means that the trajectory sequence of students with lower GPA is less similar than the other students. The results of this study indicate that trajectory information based on spatial data is related to characteristics such as student academic achievement, and it is possible to analyze characteristics of students through spatial trajectory sequence information.

Original languageEnglish
Title of host publicationProceedings of the 2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018
EditorsAkihito Sudo, Lau Hoong Chin, Takahiro Yabe, Xuan Song, Yoshihide Sekimoto
PublisherAssociation for Computing Machinery, Inc
Pages1-7
Number of pages7
ISBN (Electronic)9781450360425
DOIs
Publication statusPublished - 2018 Nov 6
Event2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018 - Seattle, United States
Duration: 2018 Nov 6 → …

Publication series

NameProceedings of the 2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018

Conference

Conference2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018
CountryUnited States
CitySeattle
Period18/11/6 → …

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Computer Science Applications
  • Control and Systems Engineering
  • Transportation

Cite this

Ju, S., Park, S., Lim, H., Yun, S. B., & Heo, J. (2018). Spatial-data-driven student characterization: Trajectory sequence alignment based on student smart card transactions. In A. Sudo, L. H. Chin, T. Yabe, X. Song, & Y. Sekimoto (Eds.), Proceedings of the 2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018 (pp. 1-7). (Proceedings of the 2nd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility, PredictGIS 2018). Association for Computing Machinery, Inc. https://doi.org/10.1145/3283590.3283591