Feature Interactions Research Articles

An adaptive chatter detection method for micro milling based on variational mode extraction

Current chatter detection methods using variational mode decomposition (VMD) hardly select appropriate intrinsic mode functions (IMFs) for characterizing chatter in micro milling. This article introduces a novel variational mode extraction (VME) method to isolate a single IMF representing chatter. An adaptive filter reduces chatter-independent components, enabling VME to identify the desired IMF with adaptive calculations for center frequency and penalty factor. The Hilbert–Huang transform is then applied to obtain the time–frequency distribution of the obtained IMF, and various features are computed. Finally, two methods are employed: streaming feature selection considers feature interaction, and radial basis function support vector machine selects pertinent features to establish the chatter detection model. Extensive micro milling tests validate the method, achieving an average detection accuracy of 99.34%.

Face forgery video detection based on expression key sequences

In order to minimize additional computational costs in detecting forged videos, and enhance detection accuracy, this paper employs dynamic facial expression sequences as key sequences, replacing original video sequences as inputs for the detection model. A spatio-temporal dual-branch detection network is designed based on the visual Transformer architecture. Specifically, this process involves three steps. Firstly, dynamic facial expression sequences are localized as key sequences using optical flow difference algorithms. Subsequently, the spatial branch network employs the focal self-attention mechanism to focus on dynamic features of expression-relevant regions and uses Factorization Machines to facilitate feature interaction among multiple key sequences. Meanwhile, the temporal branch network concentrates on learning the temporal inconsistency of optical flow differences between adjacent frames. Finally, a binary classification linear SVM combines the Softmax values from the two branch networks to provide the ultimate detection outcome. Experimental results on the Faceforensics++ dataset demonstrate: (a) replacing whole video sequences with facial expression key sequences effectively reduces training and detection time by nearly 80% and 90%, respectively; (b) compared to state-of-the-art methods involving random sequence/frame extraction and key frame extraction based on video compression techniques, the proposed approach in this paper presents a more competitive detection accuracy.

A two-stage image enhancement and dynamic feature aggregation framework for gastroscopy image segmentation

Accurate and reliable automatic segmentation of lesion areas in gastroscopy images can assist endoscopists in making diagnoses and reduce the possibility of missed or incorrect diagnoses. This paper presents a two-stage framework for segmenting gastroscopy images, which aims to improve the accuracy of medical image segmentation tasks using limited datasets. The proposed framework consists of two stages: the Image Enhancement Stage and the Lesion Segmentation Stage. First, in the Image Enhancement Stage, an image enhancement solution called TDC-Enhance is proposed to enrich the original small-scale gastroscopy image dataset. This solution performs Texture Enhancement, Detail Enhancement, and Color Enhancement on the original images. Then, in the Lesion Segmentation Stage, a multi-path automatic segmentation network for gastroscopy images, named DynaSiam, is introduced. DynaSiam comprises a Dependent Encoder, a Shared Encoder, and a Fusion Decoder. It learns feature information related to the lesion region by encoding the different enhanced images obtained in the Image Enhancement Stage as inputs to the multi-path network. Additionally, a Dynamic Feature Interaction (DFI) block is designed to capture and learn deeper image information, thereby improving the segmentation performance of the model. The experimental results show that the proposed method achieves a 90.80% mIoU, 92.71% Dice coefficient and 96.31% Accuracy. Other performance metrics also indicate the best performance, suggesting that the proposed model has significant potential for clinical analysis and diagnosis. Code and implementation details can be found on GitHub: https://github.com/kyasulee/DynaSiam.

Dual path features interaction network for efficient image super-resolution

Image super-resolution (SR) is a crucial task in computer vision that involves reconstructing a low-resolution (LR) image into its high-resolution (HR) counterpart. Transformer-based methods excel at establishing long-range dependency but face challenges with high-complexity computations and capturing fine-grained features. Conversely, CNN-based methods are advantageous in capturing fine-grained features but are limited by fixed receptive fields. Additionally, most SR methods suffer from channel redundancy, leading to higher computational overhead. In this paper, we propose a Dual Path Features Interaction Network (DPFINet) to achieve efficient image SR, which consists of two components: a) To alleviate the issue of feature channel redundancy, a Local–Global Features Modeling (LGFM) method is newly proposed, which concurrently models global and local features by splitting features along different channels. In LGFM, a Shift Window Linear Attention (SWLA) layer is adopted to effectively capture global information through a large shift window based on the split features. Meanwhile, a Multi-Scale Detail Enhancement (MSDE) layer is designed, where the split other features are encoded to facilitate detail reconstruction through an interactive fusion of semantic and local information, thereby addressing the limitations of SWLA in capturing fine-grained features. b) A Cross-Level Features Interaction (CLFI) method is proposed to fuse global and local features modeled by different network structures (SWLA and MSDE), where a novel residual fusion mechanism is designed to preserve both global and local information while complementing each other. Extensive experiments demonstrate that our method outperforms most state-of-the-art SR methods on five benchmark datasets. Notably, during inference, our approach improves performance by 0.41 dB and reduces memory consumption by approximately 79% compared to DiVANet (Behjati et al., 2023) on the Manga109 (×4) dataset.

A multidirectional pairwise coupling approach with spectral features unmixing to quantify total phosphorus, total nitrogen, and chlorophyll-a in urban rivers

Comprehensive and effective water quality monitoring is vital to water environment management and prevention of water quality from degradation. Unmanned aerial vehicle (UAV) remote sensing techniques have gradually matured and prevailed in monitoring water quality of urban rivers, posing great opportunity for more effective and flexible quantitative estimation of water quality parameter (WQP) than satellite remote sensing techniques. However, current UAV remote sensing methods often entail large quantities of cost-prohibitive in-situ collected training samples with corresponding chemical analysis in different monitoring watersheds, laying time and fiscal pressure on local environmental protection department. They suffer relatively low calculation accuracy and stability and their applicability in various watersheds is constrained. This study developed a unified two-stage method, multidirectional pairwise coupling (MDPC) with information sharing and delivery of different modeling stages to efficiently predict concentrations of WQPs including total phosphorus (TP), total nitrogen (TN), and chlorophyll-a (Chl-a) from hyperspectral data. MDPC incorporates exterior and interior feature interaction and gravity model variant to improve prediction accuracy and stability with consideration of mutual effect in the proximity. The structure design and workflow of MDPC ensure high robustness and application prospect due to achievement of good performance with less training samples, improving applicability and feasibility. The experiments show that MDPC has achieved good performance on retrieval of WQPs concentrations including TP, TN, and Chl-a, the results mean absolute percent error (MAPE) and coefficient of determination (R2) ranging from 6.34 % to 11.94 % and from 0.74 to 0.93. This study provides a systematic and scientific reference to formulate a feasible and efficient water environment management scheme.

A CNN- and Transformer-Based Dual-Branch Network for Change Detection with Cross-Layer Feature Fusion and Edge Constraints

Change detection aims to identify the difference between dual-temporal images and has garnered considerable attention over the past decade. Recently, deep learning methods have shown robust feature extraction capabilities and have achieved improved detection results; however, they exhibit limitations in preserving clear boundaries for the identified regions, which is attributed to the inadequate contextual information aggregation capabilities of feature extraction, and fail to adequately constrain the delineation of boundaries. To address this issue, a novel dual-branch feature interaction backbone network integrating the CNN and Transformer architectures to extract pixel-level change information was developed. With our method, contextual feature aggregation can be achieved by using a cross-layer feature fusion module, and a dual-branch upsampling module is employed to incorporate both spatial and channel information, enhancing the precision of the identified change areas. In addition, a boundary constraint is incorporated, leveraging an MLP module to consolidate fragmented edge information, which increases the boundary constraints within the change areas and minimizes boundary blurring effectively. Quantitative and qualitative experiments were conducted on three benchmarks, including LEVIR-CD, WHU Building, and the xBD natural disaster dataset. The comprehensive results show the superiority of the proposed method compared with previous approaches.

Combinatorial machine learning approaches for high-rise building cost prediction and their interpretability analysis

ABSTRACT This study focuses on addressing housing gaps in populous southern countries by emphasizing the importance of accurate early cost estimation in construction projects. It compares individual cost prediction models (Decision Tree, BP Neural Network, and Support Vector Machine) with combined prediction models (BP-DT and BP-SVM) for high-rise building cost prediction. The study employs interpretative analysis methods such as the Shapley additive explanation method, partial dependence plots, and individual conditional expectation plots to explore feature interplay and improve model transparency. Results show that the BP-SVM combination algorithm significantly reduces Mean Absolute Error (MAE) compared to other models. Key factors influencing cost estimation include “Estimated project duration” and “Ground floor area”, while others like “Below-ground floors” have minimal impact. The study highlights the interaction of various feature groups on prediction outcomes. Through feature interaction analysis, it is found that the Estimated project duration and Building architecture can have some negative correlation interaction. Estimated project duration and Interior decoration may have to cancel effects Overall, it contributes to construction management by aiding investors in better-assessing project profitability, thereby enhancing investment decision efficiency and quality.

Click-through rate prediction model based on graph networks and feature squeeze-and-excitation mechanism

PurposePredicting a user’s click-through rate on an advertisement or item often uses deep learning methods to mine hidden information in data features, which can provide users with more accurate personalized recommendations. However, existing works usually ignore the problem that the drift of user interests may lead to the generation of new features when they compute feature interactions. Based on this, this paper aims to design a model to address this issue.Design/methodology/approachFirst, the authors use graph neural networks to model users’ interest relationships, using the existing user features as the node features of the graph neural networks. Second, through the squeeze-and-excitation network mechanism, the user features and item features are subjected to squeeze operation and excitation operation, respectively, and the importance of the features is adaptively adjusted by learning the channel weights of the features. Finally, the feature space is divided into multiple subspaces to allocate features to different models, which can improve the performance of the model.FindingsThe authors conduct experiments on two real-world data sets, and the results show that the model can effectively improve the prediction accuracy of advertisement or item click events.Originality/valueIn the study, the authors propose graph network and feature squeeze-and-excitation model for click-through rate prediction, which is used to dynamically learn the importance of features. The results indicate the effectiveness of the model.

CFINet: Cross-Modality MRI Feature Interaction Network for Pseudoprogression Prediction of Glioblastoma.

Pseudoprogression (PSP) is a related reaction of glioblastoma treatment, and misdiagnosis can lead to unnecessary intervention. Magnetic resonance imaging (MRI) provides cross-modality images for PSP prediction studies. However, how to effectively use the complementary information between the cross-modality MRI to improve PSP prediction is still a challenging task. To address this challenge, we propose a cross-modality feature interaction network for PSP prediction. Firstly, we propose a triple-branch multi-scale module to extract low-order feature representations and a skip-connection multi-scale module to extract high-order feature representations. Then, a cross-modality interaction module based on attention mechanism is designed to make the complementary information between cross-modality MRI fully interact. Finally, the high-order cross-modality interaction information is fed into a multi-layer perceptron to achieve the PSP prediction task. We evaluate the proposed network on a private dataset with 52 subjects from Hunan Cancer Hospital and validate it on a private dataset with 30 subjects from Xiangya Hospital. The accuracy of our proposed network on the datasets is 0.954 and 0.929, respectively, which is better than most typical convolutional neural network and interaction methods.

Hierarchical channel-spatial interaction network for partial-to-partial point cloud registration

ABSTRACT Partial-to-partial point cloud registration is a fundamental and challenging task in the three-dimensional computer vision. In this letter, we propose a novel Hierarchical Channel-Spatial Interaction Network (HCSINet), which can hierarchically explore information interaction in terms of the channel and spatial dimensions, enabling well partial point cloud perception for partial-to-partial registration. Specifically, the Channel-wise Feature Interaction (CFI) module is regarded as the basic unit of the encoder sub-network, which achieves feature hierarchical interaction to learn the overlapping scores and intuitively improve the importance of overlapping areas. Furthermore, to improve the discrimination of the overlapping points, the Co-Attention-based Spatial Association (CA-SA) module progressively excavates multi-scale spatial correlation of overlapping points by conjointly computing collaborative attention. Experimental results verify the validity of our HCSINet, which achieves competitive partial-to-partial registration performances compared to existing traditional and deep learning-based methods. The source code is publicly available https://github.com/lxnudt/HCSINet.

Efficient cross-modality feature interaction for multispectral armored vehicle detection

Detecting armed vehicles from a UAV platform is challenging due to the complexity of ground environment. This paper presents a dual-stream multispectral armored vehicle detection method to tackle this problem. First, considering that there is a paucity of datasets containing multispectral armored vehicle images, a multispectral armored vehicle detection dataset is constructed for this study. The dataset consists of 5853 pairs of RGB and infrared images, featuring a total of 15,878 instances of armored vehicles. Then, a cross-modal feature interaction module is designed to enable efficient feature interaction between multispectral images. This module uses the cross-modal channel-wise feature difference method to model the channel differences between the two modal features and obtains the cross-modal channel difference matrix. The cross-modal channel difference matrix is then employed to extract the unique features of the two modal features, allowing for efficient cross-modal feature interaction by complementing each other's unique features. Experiment results demonstrate that the proposed model has excellent detection performance and is capable of coping with various challenges brought by complex ground environments.

VisioSignNet: A Dual-Interactive Neural Network for enhanced traffic sign detection

Effective traffic sign detection is crucial for the safety and operational efficiency of autonomous vehicle navigation systems, particularly in dynamically changing environments. Addressing the primary challenges of long-range pixel dependencies and enhancing the detectability of small objects in complex scenes, we present VisioSignNet: A Dual-Interactive Neural Network designed for enhanced traffic sign detection. This architecture incorporates Local and Global Interactive Modules (LGIM) and Enhancing Channel and Space Interaction (ECSI) modules. The LGIM is engineered to balance local and global feature interactions, while the ECSI optimizes the interchange of information across channel and spatial dimensions. Their synergistic interaction not only enhances the perceptual field at early processing stages but also significantly improves the recognition of small-scale, critical traffic signs. Evaluated on the TT100K and GTSDB datasets, VisioSignNet achieved mean average precision (mAP) scores of 90.5% and 97%, respectively, with a model size of 26M parameters. Its enhanced variant, VisioSignNet_l, with 34M parameters, reached mAP scores of 93.2% and 97.8%. These outcomes substantiate VisioSignNet’s efficacy in tackling the complexities of traffic sign detection, confirming its potential as a robust solution in the field of autonomous driving technologies.

Using Behavioural Biometrics and Machine Learning in Smart Gadgets for Continuous User Authentication Purposes

In the ever-evolving realm of technology, the identification of human activities using intelligent devices such as smartwatches, fitness bands, and smartphones has emerged as a crucial area of study. These devices, equipped with inertial sensors, gather a wealth of data and provide insights into users' movements and behaviors. These data not only serve practical purposes, but also hold significant implications for domains such as healthcare and fitness tracking. Traditionally, these devices have been employed to monitor various health metrics such as step counts, calorie expenditure, and real-time blood pressure monitoring. However, recent research has shifted its focus to leveraging the data collected by these sensors for user authentication purposes. This innovative approach involves the utilization of Machine Learning (ML) models to analyze the routine data captured by sensors in smart devices employing ML algorithms, which can recognize and authenticate users based on their unique movement patterns and behaviors. This introduces a paradigm shift from traditional one-time authentication methods to continuous authentication, adding an extra layer of security to protect users against potential threats. Continuous authentication offers several advantages over its conventional counterparts. First, it enhances security by constantly verifying a user's identity through their interaction with the device, thereby mitigating the risk of unauthorized access. Second, it provides a seamless and nonintrusive user experience, eliminating the need for repetitive authentication prompts. Moreover, it offers robust protection against various threats such as identity theft, unauthorized access, and device tampering. The application of continuous authentication extends beyond individual devices and encompasses interconnected systems and networks. This holistic approach ensures a comprehensive security across digital platforms and services. The experiments demonstrate that the logistic regression model achieves an accuracy of 82.32% on the test dataset, highlighting its robustness for binary classification tasks. Additionally, the random forest model outperforms with a 92.18% accuracy, emphasizing its superior capability in handling complex feature interactions. In the study, the sequential neural network achieved an accuracy of 92% on the HAR dataset, outperforming traditional machine learning models by a significant margin. The model also demonstrated robust generalization capabilities with a minimal drop in performance across various cross-validation folds.

The obstacle race to educational improvement: governance, policies, and practices in disadvantaged schools

Contemporary educational reforms emphasise school autonomy, performance-based management, and accountability as necessary policies for attaining improvement goals. Despite their widespread adoption, the configuration and enactment of these policies are discretionary and contingent, deeply influenced by the interaction of local contexts, organisational features, and teachers’ beliefs about policy and pedagogy. Apart from these challenges, schools serving disadvantaged students, often targeted for educational improvement, also struggle with significant organisational constraints and policy overload. This study explores how disadvantaged urban schools make sense of their policy environment and engage with the school improvement mandate through different logics and practices. The analytical framework stresses the policy filtering and adaptation processes that school actors undergo when tasked with multiple implementation demands, and how these interact with school contexts to produce diverse responses. This research is conducted in Catalonia, Spain, and adopts a sequential mixed-methods strategy, relying on interviews with principals and teachers, as well as survey and administrative data. The findings are organised in four ideal types of school responses to the improvement mandate, showing the diverging ways schools engage with policy. Disadvantaged schools exhibit diverse improvement approaches despite sharing structural characteristics, explained by the interaction of contextual factors and organisational characteristics. Each ideal type prioritises different aspects, such as performance, innovation, or inclusive practices. The heterogeneity of school responses highlights the necessity for targeted, context-sensitive support to ensure equitable educational practices.

Promoting low carbon construction using alkali-activated materials: A modeling study for strength prediction and feature interaction

Abstract In place of Portland cement concrete, alkali-activated materials (AAMs) are becoming more popular because of their widespread use and low environmental effects. Unfortunately, reliable property predictions have been impeded by the restrictions of conventional materials science methods and the large compositional variability of AAMs. A support vector machine (SVM), a bagging regressor (BR), and a random forest regressor (RFR) were among the machine learning models developed in this study to assess the compressive strength (CS) of AAMs in an effort to gain an answer to this topic. Improving predictions in this crucial area was the goal of this study, which used a large dataset with 381 points and eight input factors. Also, the relevance of contributing components was assessed using a shapley additive explanations (SHAP) approach. In terms of predicting AAMs CS, RFR outperformed BR and SVM. Compared to the RFR model’s 0.96 R 2, the SVM and BR models’ R 2-values were 0.89 and 0.93, respectively. In addition, the RFR model’s greater accuracy was indicated by an average absolute error value of 4.08 MPa compared to the SVM’s 6.80 MPa and the BR’s 5.83 MPa, which provided further proof of their validity. According to the outcomes of the SHAP research, the two factors that contributed the most beneficially to the strength were aggregate volumetric ratio and reactivity. The factors that contributed the most negatively were specific surface area, silicate modulus, and sodium hydroxide concentration. Using the produced models to find the CS of AAMs for various input parameter values can help cut down on costly and time-consuming laboratory testing. In order to find the best amounts of raw materials for AAMs, academics and industries could find this SHAP study useful.

Efficient and Intelligent Feature Selection via Maximum Conditional Mutual Information for Microarray Data

The challenge of analyzing microarray datasets is significantly compounded by the curse of dimensionality and the complexity of feature interactions. Addressing this, we propose a novel feature selection algorithm based on maximum conditional mutual information (MCMI) to identify a minimal feature subset that is maximally relevant and non-redundant. This algorithm leverages a greedy search strategy, prioritizing both feature quality and classification performance. Experimental results on high-dimensional microarray datasets demonstrate our algorithm’s superior ability to reduce dimensionality, eliminate redundancy, and enhance classification accuracy. Compared to existing filter feature selection methods, our approach exhibits higher adaptability and intelligence.

Inventory Prediction Using a Modified Multi-Dimensional Collaborative Wrapped Bi-Directional Long Short-Term Memory Model

Inventory prediction is concerned with the forecasting of future demand for products in order to optimize inventory levels and supply chain management. The challenges include demand volatility, data quality, multi-dimensional interactions, lead time variability, seasonal trends, and dynamic pricing. Nevertheless, these models suffer from numerous shortcomings, and in this research, we propose a new model, MMCW-BiLSTM (modified multi-dimensional collaboratively wrapped BiLSTM), for inventory prediction. The MMCW-BiLSTM model reflects a considerable leap in inventory forecasting by combining a number of components in order to consider intricate temporal dependencies and incorporate feature interactions. The MMCW-BiLSTM makes use of BiLSTM layers, collaborative attention mechanisms, and a multi-dimensional attention approach to learn from augmented datasets consisting of the original features and the extracted time series data. Moreover, adding a Taylor series transformation allows for a more precise description of the features in the model, thus improving the prediction precision. The results show that the models make the least mistakes when they use the AV demand forecasting dataset, with MAE values of 1.75, MAPE values of 2.89, MSE values of 6.76, and RMSE values of 2.6. Similarly, when utilizing the product demand dataset, the model also achieves the lowest error values for these metrics at 1.97, 3.91, 8.76, and 2.96. Likewise, when utilizing the dairy goods sales dataset, the model also achieves the lowest error values for these metrics at 2.54, 3.69, 10.39, and 3.22.

An interpretable probabilistic prediction algorithm for shield movement performance

Total thrust and torque are two key indicators of shield movement performance. Most existing data-driven machine learning studies focus on developing more accurate models for predicting total thrust and torque but overlook the interpretability of the models. To address this black-box issue, this study proposes an interpretable probabilistic prediction algorithm for the shield movement performance. The algorithm uses the natural gradient boosting (NGBoost) model to iteratively update the parametric probability distributions (e.g., mean and variance) and achieve probabilistic predictions of the total thrust and torque. The impact of each feature on the prediction values and uncertainty is quantified by extending the importance analysis of a single deterministic predictive value to both the mean and variance. The feature interactions are analyzed and their predictive contributions are quantified by the shapley additive explanations (SHAP) method. The transparency of the NGBoost model is improved through the visualization of the decision-making process. A shield tunneling project in Hangzhou is used to validate the effectiveness of the proposed algorithm. The results indicate that the NGboost model outperforms other five models in terms of accuracy. The prediction results are interpretable, and the interpretable probabilistic model provides decision-makers with a more intuitive and reliable reference.

Small Insulator Defects Detection Based on Multiscale Feature Interaction Transformer for UAV-Assisted Power IoVT

Small Insulator Defects Detection Based on Multiscale Feature Interaction Transformer for UAV-Assisted Power IoVT

Research on green supply chain finance risk identification based on two-stage deep learning

As a resonance product between financial services and the upgrading of the green industry, green supply chain finance has garnered extensive attention in the process of ecological civilization construction. Effectively promoting the green transformation of small and medium-sized enterprises and achieving the "dual carbon" goals necessitate the avoidance of corporate green risks. However, the complex interdependence and information asymmetry among green supply chain finance enterprises result in data characteristics such as multi-source small samples and high-dimensional imbalance. To address these issues, this paper proposes a risk assessment model based on two-stage deep learning. In the first stage, we employ Generative Adversarial Network (GAN) to generate minority class default samples, and utilize Stacked Auto-Encoder (SAE) to extract data features with closed-form parameter calculation capability. In the second stage, the obtained features are input into a Deep Neural Network (DNN), and parameter learning and model optimization are conducted through joint training. Finally, to model low-order feature interactions, we integrate the Support Vector Machine (SVM) algorithm. The paper is grounded in the green innovation production of enterprises, collecting financial data of 176 upstream and downstream enterprises and corresponding core enterprise green indicators from 2013 to 2022. Experimental results demonstrate that GAN oversampling technique not only enhances the model's AUC metric but also significantly improves the F1 score. Compared with traditional deep learning methods, the proposed two-stage deep integration model effectively reduces training loss and exhibits superiority in identifying green supply chain finance risks.