English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Prediction tasks about students have practical significance for both studentand college. Making multiple predictions about students is an important part ofa smart campus. For instance, predicting whether a student will fail tograduate can alert the student affairs office to take predictive measures tohelp the student improve his/her academic performance. With the development ofinformation technology in colleges, we can collect digital footprints whichencode heterogeneous behaviors continuously. In this paper, we focus onmodeling heterogeneous behaviors and making multiple predictions together,since some prediction tasks are related and learning the model for a specifictask may have the data sparsity problem. To this end, we propose a variant ofLSTM and a soft-attention mechanism. The proposed LSTM is able to learn thestudent profile-aware representation from heterogeneous behavior sequences. Theproposed soft-attention mechanism can dynamically learn different importancedegrees of different days for every student. In this way, heterogeneousbehaviors can be well modeled. In order to model interactions among multipleprediction tasks, we propose a co-attention mechanism based unit. With the helpof the stacked units, we can explicitly control the knowledge transfer amongmultiple tasks. We design three motivating behavior prediction tasks based on areal-world dataset collected from a college. Qualitative and quantitativeexperiments on the three prediction tasks have demonstrated the effectivenessof our model.

Jointly Modeling Heterogeneous Student Behaviors and Interactions Among Multiple Prediction Tasks