Multivariate time series classification (MTSC) is a significant research topic in the realm of data mining, with broad applications in different industries, including healthcare, finance, meteorology, and traffic. While existing studies have designed many classifiers based on LSTMs, CNNs, and Transformer, the sophisticated architectures raise concerns regarding efficiency in computation. Additionally, most methods concentrate on a single dimension, typically temporal patterns, without fully considering multi-dimensional information such as the independence and interactions across variables that are essential in multivariate settings. To address these challenges, this article introduces FreConvNet, a lightweight convolutional network integrated with frequency adaptive learning. Inheriting the modular design paradigm of Transformer to achieve multi-view modeling of multivariate time series. FreConvNet consists of two key components: the frequency adaptive block (FAB) and the convolutional feed-forward network (ConvFFN). The FAB leverages the Fourier Transform in conjunction with adaptive filters to capture both long-term and short-term dependencies in the temporal dimension. Following that, ConvFFN captures cross-variable and cross-feature interactions by controlling inter-channel information flow through grouped pointwise convolutions, while introducing non-linearity to enhance representational capacity. Extensive experiments conducted on the well-known UEA archive validate that FreConvNet outperforms existing convolution-based, Transformer-based, and hybrid methods in classification performance and offers a computationally efficient solution.

Hashing techniques have attracted considerable attention owing to their advantages of efficient computation and economical storage. However, it is still a challenging problem to generate more compact binary codes for promising performance. In this paper, we propose a novel contrastive vision transformer hashing method, which seamlessly integrates contrastive learning and vision transformers (ViTs) with hash technology into a well-designed model to learn informative features and compact binary codes simultaneously. First, we modify the basic contrastive learning framework by designing several hash layers to meet the specific requirement of hash learning. In our hash network, ViTs are applied as backbones for feature learning, which is rarely performed in existing hash learning methods. Then, we design a multiobjective loss function, in which contrastive loss explores discriminative features by maximizing agreement between different augmented views from the same image, similarity preservation loss performs pairwise semantic preservation to enhance the representative capabilities of hash codes, and quantization loss controls the quantitative error. Hence, we can facilitate end-to-end joint training to improve the retrieval performance. The encouraging experimental results on three widely used benchmark databases demonstrate the superiority of our algorithm compared with several state-of-the-art hashing algorithms.

Due to the progressive complexity of traffic networks, the traffic pressure increases sharply and traffic accidents occur frequently, which is largely posed by imprecise traffic information service provided in existing Intelligent traffic service systems (ITSSs). To relieve traffic information trek encountered in complex traffic networks, the personalized traffic information recommendation based on click through rate (CTR) prediction has attracted extensive attention. However, the data sparsity and cold start problems in traditional recommendations hinder their real-time applications in ITSSs. In this paper, the multisource data with different types of context information are utilized to construct a personalized fine-grained recommendation method for intelligent traffic services (PF-ITS), which includes three components: a road condition optimization strategy (RCOS) to capture users' behavior preferences and traffic patterns, an encoder–decoder long short-term memory (LSTM) model to address the data sparsity and cold start problems with rich context information, and an improved DeepFM model based on attention mechanism (DeepAFM) to exert personalized fine-grained traffic information recommendation with embeddings of multisource data. More specifically, the RCOS is proposed to perform coarse-grained route recommendation based on path planning theory, in which the traffic data and drivers' preferences are fully utilized for comprehensive modeling. The encoder–decoder LSTM model is employed for representation learning, in which the traffic sequences and driving behaviors are mapped into dense distributed representations with rich semantic information. The DeepAFM is utilized to achieve effective driving safety guarantee according to an individual's requirements, in which the weighted low-order feature combinations and high-order feature interactions are incorporated for personalized fine-grained recommendation. We also conduct extensive experiments on public data sets and in real-world scenarios. The experimental results demonstrate that the PF-ITS method based on RCOS and DeepAFM outperforms the state-of-the-art baseline models in effectiveness and efficiency.

Emotion recognition has become a research focus in the brain–computer interface and cognitive neuroscience. Electroencephalogram (EEG) is employed for its advantages as accurate, objective, and noninvasive nature. However, many existing research only focus on extracting the time and frequency domain features of the EEG signals while failing to utilize the dynamic temporal changes and the positional relationships between different electrode channels. To fill this gap, we develop the dynamic differential entropy and brain connectivity features based EEG emotion recognition using linear graph convolutional network named DDELGCN. First, the dynamic differential entropy feature which represents the frequency domain feature as well as time domain feature is extracted based on the traditional differential entropy feature. Second, brain connectivity matrices are constructed by calculating the Pearson correlation coefficient, phase-locked value and transfer entropy, and then are used to denote the connectivity features of all electrode combinations. Finally, a linear graph convolutional network is customized and applied to aggregate the features from total electrode combinations and then classifies the emotional states, which consists of five layers, namely, an input layer, two linear graph convolutional layers, a fully connected layer, and a softmax layer. Extensive experiments show that the accuracies in the valence and arousal dimensions reach 90.88% and 91.13%, and the precision reaches 96.66% and 97.02% on the DEAP dataset, respectively. On the SEED dataset, the accuracy and precision reach 91.56% and 97.38%, respectively.

Emotion recognition has become an important component of human–computer interaction systems. Research on emotion recognition based on electroencephalogram (EEG) signals are mostly conducted by the analysis of all channels' EEG signals. Although some progresses are achieved, there are still several challenges such as high dimensions, correlation between different features and feature redundancy in the realistic experimental process. These challenges have hindered the applications of emotion recognition to portable human–computer interaction systems (or devices). This paper explores how to find out the most effective EEG features and channels for emotion recognition so as to only collect data as less as possible. First, discriminative features of EEG signals from different dimensionalities are extracted for emotion classification, including the first difference, multiscale permutation entropy, Higuchi fractal dimension, and discrete wavelet transform. Second, relief algorithm and floating generalized sequential backward selection algorithm are integrated as a novel channel selection method. Then, support vector machine is employed to classify the emotions for verifying the performance of the channel selection method and extracted features. At last, experimental results demonstrate that the optimal channel set, which are mostly located at the frontal, has extremely high similarity on the self-collected data set and the public data set and the average classification accuracy is achieved up to 91.31% with the selected 10-channel EEG signals. The findings are valuable for the practical EEG-based emotion recognition systems.

In recent years, automatic emotion recognition renders human–computer interaction systems intelligent and friendly. Emotion recognition based on electroencephalogram (EEG) has received widespread attention and many research results have emerged, but how to establish an integrated temporal and spatial feature fusion and classification method with improved convolutional neural networks (CNNs) and how to utilize the spatial information of different electrode channels to improve the accuracy of emotion recognition in the deep learning are two important challenges. This paper proposes an emotion recognition method based on three-dimensional (3D) feature maps and CNNs. First, EEG data are calibrated with 3 s baseline data and divided into segments with 6 s time window, and then the wavelet energy ratio, wavelet entropy of five rhythms, and approximate entropy are extracted from each segment. Second, the extracted features are arranged according to EEG channel mapping positions, and then each segment is converted into a 3D feature map, which is used to simulate the relative position of electrode channels on the scalp and provides spatial information for emotion recognition. Finally, a CNN framework is designed to learn local connections among electrode channels from 3D feature maps and to improve the accuracy of emotion recognition. The experiments on data set for emotion analysis using physiological signals data set were conducted and the average classification accuracy of 93.61% and 94.04% for valence and arousal was attained in subject-dependent experiments while 83.83% and 84.53% in subject-independent experiments. The experimental results demonstrate that the proposed method has better classification accuracy than the state-of-the-art methods.