Intelligent learning diagnosis is a critical engine of intelligent tutoring systems, which aims to estimate learners’ current knowledge mastery status and predict their future learning performance. The significant challenge with traditional learning diagnosis methods is the inability to balance diagnostic accuracy and interpretability. Although the existing psychometric‐based learning diagnosis methods provide some domain interpretation through cognitive parameters, they have insufficient modeling capability with a shallow structure for large‐scale learning data. While the deep learning‐based learning diagnosis methods have improved the accuracy of learning performance prediction, their inherent black‐box properties lead to a lack of interpretability, making their results untrustworthy for educational applications. To settle the abovementioned problem, the proposed unified interpretable intelligent learning diagnosis framework, which benefits from the powerful representation learning ability of deep learning and the interpretability of psychometrics, achieves a better performance of learning prediction and provides interpretability from three aspects: cognitive parameters, learner‐resource response network, and weights of self‐attention mechanism. Within the proposed framework, this paper presents a two‐channel learning diagnosis mechanism LDM‐ID as well as a three‐channel learning diagnosis mechanism LDM‐HMI. Experiments on two real‐world datasets and a simulation dataset show that our method has higher accuracy in predicting learners’ performances compared with the state‐of‐the‐art models and can provide valuable educational interpretability for applications such as precise learning resource recommendation and personalized learning tutoring in intelligent tutoring systems.

A robust and scientifically grounded teaching evaluation system holds significant importance in modern education, serving as a crucial metric that reflects the quality of classroom instruction. However, current methodologies within smart classroom environments have distinct limitations. These include accommodating a substantial student population, grappling with object detection challenges due to obstructions, and encountering accuracy issues in recognition stemming from varying observation angles. To address these limitations, this paper proposes an innovative data augmentation approach designed to detect distinct student behaviors by leveraging focused behavioral attributes. The primary objective is to alleviate the pedagogical workload. The process begins with assembling a concise dataset tailored for discerning student learning behaviors, followed by the application of data augmentation techniques to significantly expand its size. Additionally, the architectural prowess of the Extended-efficient Layer Aggregation Networks (E-ELAN) is harnessed to effectively extract a diverse array of learning behavior features. Of particular note is the integration of the Channel-wise Attention Module (CBAM) focal mechanism into the feature detection network. This integration plays a pivotal role, enhancing the network’s ability to detect key cues relevant to student learning behaviors and thereby heightening feature identification precision. The culmination of this methodological journey involves the classification of the extracted features through a dual-pronged conduit: the Feature Pyramid Network (FPN) and the Path Aggregation Network (PAN). Empirical evidence vividly demonstrates the potency of the proposed methodology, yielding a mean average precision (mAP) of 96.7%. This achievement surpasses comparable methodologies by a substantial margin of at least 11.9%, conclusively highlighting the method’s superior recognition capabilities. This research has an important impact on the field of teaching evaluation system, which helps to reduce the burden of educators on the one hand, and makes teaching evaluation more objective and accurate on the other hand.

Digital audio tampering detection can be used to verify the authenticity of digital audio. However, most current methods use standard electronic network frequency (ENF) databases for visual comparison analysis of ENF continuity of digital audio or perform feature extraction for classification by machine learning methods. ENF databases are usually tricky to obtain, visual methods have weak feature representation, and machine learning methods have more information loss in features, resulting in low detection accuracy. This paper proposes a fusion method of shallow and deep features to fully use ENF information by exploiting the complementary nature of features at different levels to more accurately describe the changes in inconsistency produced by tampering operations to raw digital audio. Firstly, the audio signal is band-pass filtered to obtain the ENF component. Then, the discrete Fourier transform (DFT) and Hilbert transform are performed to obtain the phase and instantaneous frequency of the ENF component. Secondly, the mean value of the sequence variation is used as the shallow feature; the feature matrix obtained by framing and reshaping of the ENF sequence is used as the input of the convolutional neural network; the characteristics of the fitted coefficients are obtained by curve fitting. Then, the local details of ENF are obtained from the feature matrix by the convolutional neural network, and the global information of ENF is obtained by fitting coefficient features through deep neural network (DNN). The depth features of ENF are composed of ENF global information and local information together. The shallow and deep features are fused using an attention mechanism to give greater weights to features useful for classification and suppress invalid features. Finally, the tampered audio is detected by downscaling and fitting with a DNN containing two fully connected layers, and classification is performed using a Softmax layer. The method achieves 97.03% accuracy on three classic databases: Carioca 1, Carioca 2, and New Spanish. In addition, we have achieved an accuracy of 88.31% on the newly constructed database GAUDI-DI. Experimental results show that the proposed method is superior to the state-of-the-art method.

Artificial intelligence (AI) and computer vision technologies have gained significant prominence in the field of education. These technologies enable the detection and analysis of students’ classroom behaviors, providing valuable insights for assessing individual concentration levels. However, the accuracy of target detection methods based on Convolutional Neural Networks (CNNs) can be compromised in classrooms with multiple targets and varying scales, as convolutional operations may result in the loss of location information. In contrast, transformers, which leverage attention mechanisms, have the capability to learn global features and mitigate the information loss caused by convolutional operations. In this paper, we propose a students’ classroom behavior detection system that combines deformable DETR with a Swin Transformer and light-weight Feature Pyramid Network (FPN). By employing a feature pyramid structure, the system can effectively process multi-scale feature maps extracted by the Swin Transformer, thereby improving the detection accuracy for targets of different sizes and scales. Moreover, the integration of the CARAFE lightweight operator into the FPN structure enhances the network’s detection accuracy. To validate the effectiveness of our approach, extensive experiments are conducted on a real dataset of students’ classroom behavior. The experimental results demonstrate a significant 6.1% improvement in detection accuracy compared to state-of-the-art methods. These findings highlight the superiority of our proposed network in accurately detecting and analyzing students’ classroom behaviors. Overall, this research contributes to the field of education by addressing the limitations of CNN-based target detection methods and leveraging the capabilities of transformers to improve accuracy. The proposed system showcases the benefits of integrating deformable DETR, Swin Transformer, and the lightweight FPN in the context of students’ classroom behavior detection. The experimental results provide compelling evidence of the system’s effectiveness and its potential to enhance classroom monitoring and assessment practices.

The automated classification of breast cancer histopathological images is one of the important tasks in computer-aided diagnosis systems (CADs). Due to the characteristics of small inter-class and large intra-class variances in breast cancer histopathological images, extracting features for breast cancer classification is difficult. To address this problem, an improved autoencoder (AE) network using a Siamese framework that can learn the effective features from histopathological images for CAD breast cancer classification tasks was designed. First, the inputted image is processed at multiple scales using a Gaussian pyramid to obtain multi-scale features. Second, in the feature extraction stage, a Siamese framework is used to constrain the pre-trained AE so that the extracted features have smaller intra-class variance and larger inter-class variance. Experimental results show that the proposed method classification accuracy was as high as 97.8% on the BreakHis dataset. Compared with commonly used algorithms in breast cancer histopathological classification, this method has superior, faster performance.