Completing missing exam scores with structural information and beyond

Xiao Jiang; Limei Zhang; Lishan Qiao

doi:10.1117/1.JRS.13.022005

10 November 2018 Completing missing exam scores with structural information and beyond

Xiao Jiang, Limei Zhang, Lishan Qiao

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 13, Issue 2, 022005 (November 2018). https://doi.org/10.1117/1.JRS.13.022005

Abstract

Score missing is a common problem involved in exam datasets due to, for example, the absence of students from a test, major-transfer, accidental deletion by operators, etc. Incomplete data tend to bring great inconvenience to analysis, comparison, and evaluation, which may further affect the reliability of the final conclusion and the subsequent decision-making (e.g., adjustment of the teaching plan). Although, in principle, most of the existing data completion methods can be directly used on the exam score dataset, to our best knowledge, there is no systematic evaluation on these methods for the score completion problem. Moreover, the general completion methods cannot effectively employ the special structural information in the score dataset (e.g., the order structure in the students and the relationship between different subjects). Therefore, (1) we conduct a comparative study on the mainstream data completion methods for estimating the missing values in score datasets, (2) we propose an easy-to-implement score completion method by explicitly using informative structures in the dataset, which achieves the best estimation/prediction accuracy with a high efficiency. Beyond the exam scores, we also verify the flexibility of our proposed method in completing remote sensing data.

Citation Download Citation

Xiao Jiang, Limei Zhang, and Lishan Qiao "Completing missing exam scores with structural information and beyond," Journal of Applied Remote Sensing 13(2), 022005 (10 November 2018). https://doi.org/10.1117/1.JRS.13.022005

Received: 3 August 2018; Accepted: 23 October 2018; Published: 10 November 2018

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available