Research On The Assessment Of Group Ability And Its Intelligent Applications

With the rapid development of educational informatization,more and more online learning platforms have been established to provide personalized and intelligent teaching services for students.In intelligent education,cognitive diagnosis is a basic and important task.Its goal is to use students’ learning behavior data to model and analyze students’ learning ability and cognitive level.Most cognitive diagnosis research in education has been concentrated on individual assessment,aiming at discovering the latent characteristics of students.However,in many real-world scenarios,group-level assessment is an important and meaningful task,e.g.,class assessment in different regions can discover the difference of teaching level in different contexts.In this work,we consider assessing cognitive ability for a group of students,so as to help explore and analyze the learning state of groups,known as the group-level cognitive diagnosis.Due to the challenges of sparse group-exercise response data,inaccurate modeling of high-order interaction,and poor interpretability of diagnostic results,the task of group-level ability assessment is still worthy of in-depth exploration.To this end,this dissertation carries out an in-depth study on the relevant challenges,and propose a general Multi-Task based Group-Level Cognitive Diagnosis(MGCD)framework.In general,the research contents and contributions of this paper are listed as follows:1)We jointly model student-exercise responses and group-exercise responses in a multi-task manner to alleviate the sparsity of group-exercise responses;2)We design a context-aware attention network to model the relationship between student knowledge state and group knowledge state in different contexts;3)We model an interpretable cognitive layer to obtain student ability,group ability and exercise factors(e.g.,difficulty),and then we leverage neural networks to learn complex interaction functions among them.Extensive experiments on real-world datasets demonstrate the generality of MGCD and the effectiveness of our attention design and multi-task learning.