Automatic short answer scoring (ASAS) helps reduce the grading burden on educators but often lacks detailed, explainable feedback. Existing methods in ASAS with feedback (ASAS-F) rely on fine-tuning language models with limited datasets, which is resource-intensive and struggles to generalize across contexts. Recent approaches using large language models (LLMs) have focused on scoring without extensive fine-tuning. However, they often rely heavily on prompt engineering and either fail to generate elaborated feedback or do not adequately evaluate it. In this paper, we propose a modular retrieval augmented generation based ASAS-F system that scores answers and generates feedback in strict zero-shot and few-shot learning scenarios. We design our system to be adaptable to various educational tasks without extensive prompt engineering using an automatic prompt generation framework. Results show an improvement in scoring accuracy by 9% on unseen questions compared to fine-tuning, offering a scalable and cost-effective solution.

While Graph Convolutional Networks (GCNs) have shown great potential in recommender systems and collaborative filtering (CF), they suffer from expensive computational complexity and poor scalability. On top of that, recent works mostly combine GCNs with other advanced algorithms which further sacrifice model efficiency and scalability. In this work, we unveil the redundancy of existing GCN-based methods in three aspects: (1) Feature redundancy . By reviewing GCNs from a spectral perspective, we show that most spectral graph features are noisy for recommendation, while stacking graph convolution layers can suppress but cannot completely remove the noisy features, which we mostly summarize from our previous work; (2) Structure redundancy . By providing a deep insight into how user/item representations are generated, we show that what makes them distinctive lies in the spectral graph features, while the core idea of GCNs (i.e., neighborhood aggregation) is not the reason making GCNs effective; and (3) Distribution redundancy . Following observations from (1), we further show that the number of required spectral features is closely related to the spectral distribution, where important information tends to be concentrated in more (fewer) spectral features on a flatter (sharper) distribution. To make important information be concentrated in as few features as possible, we sharpen the spectral distribution by increasing the node similarity without changing the original data, thereby reducing the computational cost. To remove these three kinds of redundancies, we propose a Simplified Graph Denoising Encoder (SGDE) only exploiting the top- K singular vectors without explicitly aggregating neighborhood, which significantly reduces the complexity of GCN-based methods. We further propose a scalable contrastive learning framework to alleviate data sparsity and to boost model robustness and generalization, leading to significant improvement. Extensive experiments on three real-world datasets show that our proposed SGDE not only achieves state-of-the-art but also shows higher scalability and efficiency than our previously proposed GDE as well as traditional and GCN-based CF methods.

Examining student learning behavior is one of the crucial educational issues. In this paper, we propose a new method to predict student performance by using comment data mining. A teacher just asks students after every lesson to freely describe and write about their learning situations, attitudes, tendencies, and behaviors. The method employs Latent Dirichlet Allocation (LDA) and Support Vector Machine (SVM) to predict student grades in each lesson. In order to obtain further improvement of prediction results, we apply a majority vote method to the predicted results obtained in consecutive lessons to keep track of each student's learning situation. Also, we evaluate the reliability of the predicted student grades to know when we can rely prediction results of student grade during the period of the semester. The experiment results show that our proposed method continuously tracked student learning situation and improved prediction performance of final student grades compared to Probabilistic Latent Semantic Analysis (PLSA) and Latent Semantic Analysis (LSA) models. Also, considering the differences of prediction results in the two consecutive lessons helps to evaluate the reliability of the predicted results.

Most current prediction models are difficult for teachers to interpret. This induces significant problems of grasping characteristics for each grade group of students, which are helpful for giving intervention and providing feedback to them. In this paper, we propose a new method to build a practical prediction model based on comment data mining. The current study classifies students' comments into six attributes (attitudes, finding, cooperation, review the lesson, understanding, and next activity plan), then extracts generic rules 'IF-THEN' about students' activities, attitudes and situations in the learning environment. Decision Tree (DT) and Random Forest (RF) models are applied to discriminate unique features related to each grade group. Evaluation results reported a set of rules for students' performance among with their situations reflected through all the course of a semester.