ABSTRACT Detecting student behaviour objects in classroom environments is crucial for assessing educational progress, optimizing teaching strategies and improving student learning outcomes. With the ongoing advancement of educational informatization, analysing classroom behaviour has become an important tool for enhancing teaching quality and personalized learning. However, current student behaviour object detection models based on CNN and Transformer architectures face challenges such as large parameter sizes and high inference delays when deployed on edge devices in classrooms, limiting their practical application. To address these issues, this study proposes a lightweight student behaviour detection framework based on the Mamba architecture, aimed at balancing computational efficiency and detection accuracy. First, the framework based on the state‐space model (SSM) efficiently captures global dependencies, using local convolutions to enhance detection accuracy and scene understanding while maintaining real‐time performance. Second, the C2CGA module increases attention diversity through feature splitting, self‐attention, cascading and projected concatenation, deepening the network while reducing computational overhead. Finally, the A2CMoCA module aggregates multi‐scale features, improving the learning of small objects and occluded behaviours. Experiments on a self‐built classroom behaviour dataset (containing eight typical teaching behaviours) show that the proposed method achieves 91.5% detection accuracy while maintaining a lightweight design. Compared to the baseline model, its computational efficiency (5.9G FLOPs) is reduced by 56.6%, the parameter size is compressed to 3.65 M (a 39% reduction) and the inference speed is 3.2 ms, meeting the real‐time monitoring requirements in classroom teaching scenarios.

Stack Overflow is a widely-used community Q&A website for programming-related queries.In such a platform, providing related questions as suggestions to the users can significantly enhance their search experience.Although there are many approaches based on deep learning that can automatically predict the relatedness between questions, those approaches are limited because the semantic and interaction features of the sentences may be lost.In this paper, we propose a novel method to predict the relatedness between questions based on semantic matching.We adopt the Interaction Feature Extractor to capture the interaction information and fuse it through a fusion mechanism to enhance the interaction between questions.Our experimental results demonstrate that our proposed method achieves stateof-the-art performance in terms of Precision, Recall, and F1score evaluation metrics, outperforming the baseline approaches.Furthermore, we show that our model also performs well in other semantic matching tasks in software fields, indicating its generalization ability and robustness.

As one of the most popular programming forums, Stack Overflow has helped many developers with massive high-quality questions and answers. Particularly, the related questions identified by developers can supply targeted knowledge to solve the programming problems. However, it is difficult to identify all relevant questions by developers from massive questions in Stack Overflow. Although some studies have raised methods for automatically identifying relatedness between questions, only a few of them provided related questions to new query. In addition, the existing methods can not extract the global information between query and candidate questions in a proper way. In this paper, we propose a novel method that recommends the related questions to developers' new queries based on the semantic matching. We introduce a novel integral fusion to improve the global information extraction and use the inter-attention to capture the local interactive information. Besides, we have pre-trained domain-specific word embeddings to enhance the processing of software engineering information. The experiment results show that our model achieves competitive performance in MRR, nDCG@5, and nDCG@10 metrics in the related questions retrieval on Stack Overflow,

Due to the highly competitive and dynamic mobile application (app) market, app developers need to release new versions regularly to improve existing features and provide new features for users.To accomplish the maintenance and evolution tasks more effectively and efficiently, app developers should collect and analyze user reviews, which contain a rich source of information from user perspective.Although there are many approaches based on intention mining that can automatically predict the intention of reviews for better understanding valuable information, those approaches are limited since contextual information of the whole review text may be lost.In this paper, we propose Mining Intention from App Reviews (MIAR), a novel deep learning model to predict the intention of app reviews automatically.We adopt a Contextual Feature Extractor to capture the context semantic information and fuse it with the local feature through a fusion mechanism.The experiment results demonstrate that MIAR has made significant improvement over the baseline approaches in Precision, Recall, and F1-score evaluation metrics, achieving state-of-the-art performance in this task.Our model also performs well in other intention mining tasks, proving its generalization ability and robustness.

Augmented reality head-up display (AR HUD) is a new technology in assisted driving, which can add extra information to the driving environment in real-time to help the driver better perceive road situation. AR HUD can enhance driving safety but may also encourage inattentional blindness. Hence, this study aims to examine whether AR HUD-induces inattentional blindness and determine whether workload intensifies their relationship. In experiment 1, 60 participants were randomly assigned to three groups and watched three types of augmented reality (AR)-augmented driving videos, respectively. They were instructed to respond to any critical events, but only their responses to road-crossing pedestrians were recorded. Results show that AR HUD reduces inattentional blindness when pedestrians are augmented but encourages inattentional blindness when pedestrians are not augmented. In experiment 2, 20 participants viewed AR-augmented driving videos of high and low workloads. Pedestrians were not augmented in all videos. Result reveals that a high workload induces more inattentional blindness than low workload. The finding confirms that AR HUD induces inattentional blindness, and a high workload will intensify this relationship. The future design of the AR HUD assisted-driving system should consider the risk of inattentional blindness and come up with corresponding countermeasures.

Due to the highly competitive and dynamic mobile application (app) market, app developers need to release new versions regularly to improve existing features and provide new features for users. To accomplish the maintenance and evolution tasks more effectively and efficiently, app developers should collect and analyze user reviews, which contain a rich source of information from user perspective. Although there are many approaches based on intention mining that can automatically predict the intention of reviews for better understanding valuable information, those approaches are limited since contextual information of the whole review text may be lost. In this paper, we propose Mining Intention from App Reviews (MIAR), a novel deep learning model to predict the intention of app reviews automatically. We adopt a Contextual Feature Extractor to capture the context semantic information and fuse it with the local feature through a fusion mechanism. The experiment results demonstrate that MIAR has made significant improvement over the baseline approaches in Precision, Recall and [Formula: see text]-score evaluation metrics, achieving state-of-the-art performance in this task. Our model also performs well in other intention mining tasks, proving its generalization ability and robustness.

Stack Overflow is one of the most popular Programming Community-based Question Answering (PCQA) websites that has attracted more and more users in recent years. When users raise or inquire questions in Stack Overflow, providing related questions can help them solve problems. Although there are many approaches based on deep learning that can automatically predict the relatedness between questions, those approaches are limited since interaction information between two questions may be lost. In this paper, we adopt the deep learning technique, propose an Attention-based Sentence pair Interaction Model (ASIM) to predict the relatedness between questions on Stack Overflow automatically. We adopt the attention mechanism to capture the semantic interaction information between the questions. Besides, we have pre-trained and released word embeddings specific to the software engineering domain for this task, which may also help other related tasks. The experiment results demonstrate that ASIM has made significant improvement over the baseline approaches in Precision, Recall, and Micro-F1 evaluation metrics, achieving state-of-the-art performance in this task. Our model also performs well in the duplicate question detection task of AskUbuntu, which is a similar but different task, proving its generalization and robustness.

This paper presents a high linearity dynamic amplifier with an on-chip gain detection technique, which realizes the stabilization of process, voltage, and temperature (PVT) with the real-time fluctuation. Based on the cross-coupled capacitive degeneration topology and characteristics of transistors in weak inversion, according to the simulation results, the proposed dynamic amplifier achieves -80 dB total harmonic distortion (THD) across the wide range of supply voltage and temperature. The prototype amplifier is designed at the supply voltage of 1 V in 28 nm CMOS process, and it dissipates 260 μW at the frequency of 100 MS/s with the maximum output swing of 1.2 Vpp. The power consumption is positively related to the clock frequency.

We present a PatternQuest framework to learn the patterns of interest (i.e., the distribution patterns of positive objects) using classification methods and relevance feedback. To improve the performance of multimedia retrieval, our PatternQuest first employs an efficient feature selection method to extract a low-dimensional feature subspace. With the feature selection, PatternQuest can effectively alleviate the curse of dimensionality for learning-based relevance feedback. To discover patterns of interest in the feature subspace effectively, we propose a multiresolution pattern discovery (MPD) approach, which trains an online pattern classification method known as adaptive random forests to filter negative objects, from the neighborhood of the query to the global scope, in a fine to coarse way. With MPD, our PatternQuest method can iteratively capture the patterns of interest with a little training data from the user's feedback. We have carried out extensive experiments on an image database (with 31,438 Corel images) to demonstrate the effectiveness and robustness of our method.

Relevance feedback has been an indispensable component for multimedia retrieval systems. In this paper, we present an adaptive pattern discovery method, which addresses relevance feedback by interactively discovering meaningful patterns of relevant objects. To facilitate pattern discovery, we first present a dynamic feature extraction method, which aims to alleviate the curse of dimensionality by extracting a feature subspace using balanced information gain. In the feature subspace, we train an online pattern classification method called adaptive random forests to classify multimedia objects as relevant or irrelevant. Our adaptive random forests adapts the traditional classification method known as random forests for relevance feedback. It improves the efficiency of pattern discovery by choosing the most-informative samples for online learning. Extensive experiments are carried out on a Corel image set (with 31,438 images) to evaluate the performance of our method as compared against the state-of-the-art approaches.

Relevance feedback can effectively improve the performance of content-based multimedia retrieval systems. To be effective, a relevance feedback approach must be able to efficiently capture the user's query concept from a very limited number of training samples. To address this issue, we propose a novel adaptive classification method using random forests, which is a machine learning algorithm with proven good performance on many traditional classification problems. With random forests, our method reduces the relevance feedback to a two-class classification problem and classifies database objects as relevant or irrelevant. From the relevant object set, our approach returns the top k nearest neighbors of the query to the user. Briefly speaking, our relevance feedback method has the following dominant features. First, our method is able to address the multimodal distribution of relevant points, because it trains a nonparametric and nonlinear classifier, i.e., random forests, for relevance feedback. Second, it does not overfit training data because it uses an ensemble of tree classifiers to classify multimedia objects. Experiments on a Corel image set (with 31,438 images) show that our method significantly outperforms the state-of-the-art relevance feedback approaches.

Users of image databases often prefer to retrieve relevant images by categories. Unfortunately, images are usually indexed by low-level features like color, texture and shape, which often fail to capture high-level concepts well. To address this issue, relevance feedback has been extensively used to associate low-level image features with highlevel concepts. Among all existing relevance feedback approaches, query movement and feature re-weighting have been proven to be suitable for large-scaled image databases with high dimensional image features. We present a feature re-weighting approach using relevant images as well as irrelevant ones in the relevance feedback. As far as feature re-weighting approaches are concerned, one of their common drawbacks is that the feature re-weighting process is prone to be trapped by suboptimal states. To overcome this problem, we introduce a disturbing factor, which is based on the Fisher criterion, to push the feature weights out of sub-optimum. Experimental results on a large-scaled image database with 31,438 COREL images demonstrate the effectiveness of the presented method.

We present a self-adjustable metadatabase aimed at improving the performance of the relevance feedback module extensively used in content-based image retrieval systems. Our metadatabase provides a mechanism for accumulating the optimized relevance feedback records (which are called metadata records) obtained from previous queries. Each metadata record in the metadatabase includes optimal query, feature weights, and identifiers of relevant and/or irrelevant images, and can be effectively used to guide future queries. With the metadatabase, the relevance feedback module admits a noticeable improvement on its performance for category-based search, especially when the relevant images form multiple classes in the feature space. Experiments on a Corel image set (with 31,438 images) show that our method has at least a 15% improvement on average precision and recall over relevance-feedback-only approaches.

This article introduces the basic segmentation problems in Chinese handwriting and also several prior work to solve these problems. A new segmentation method is proposed, which is applicable to both on-line and off-line systems for free-format handwritten Chinese character sentences. This method performs basic segmentation and fine segmentation based on the varying spacing thresholds and the minimum variance criteria. The five most probable ways of segmentation are derived from this stage and all the possible segments are extracted and recognized. A lattice is created from all the segments and searched using a viterbi based algorithm to find the most likely character sequence. The algorithm presented in this paper provides large flexibility and robustness to handle free-format continuous Chinese handwriting and is a promising solution for a natural and fast Chinese pen input system. The character accuracy is 85.0% for on-line and 77.4% for the off-line test data.