Mainstream Approaches Research Articles

Many-Objective Jaccard-Based Evolutionary Feature Selection for High-Dimensional Imbalanced Data Classification.

Filters and wrappers represent two mainstream approaches to feature selection (FS). Although evolutionary wrapper-based FS outperforms filters in addressing real-world classification problems, extending these methods to high-dimensional, many-objective optimization problems with imbalanced data poses substantial challenges. Overcoming computational costs and identifying suitable performance metrics are vital for navigating search operation complexities. Here, we propose using the Jaccard similarity (JS) in a set-based evolutionary many-objective (JSEMO) FS search, addressing both evolutionary FS and imbalanced classifier choice concurrently. This study highlights the mutual influence between these aspects, impacting overall algorithm performance. JSEMO integrates JS into population initialization, reproduction, and elitism steps, enhancing diversity and avoiding duplicate solutions. The set-based variation operator utilizes intersection and union operators for compatibility with binary coding. We also introduce a double-weighted KNN (KNN2W) classifier with four supportive objectives as a many-objective FS problem to handle imbalanced distributions. Compared with 20 methods on 15 benchmark problems, JSEMO produces distinct optimal features, significantly improving overall accuracy, balance accuracy, and g-mean metrics with comparable feature set size and computational cost. The ablation study underscores the positive impact of all JSEMO components, highlighting the set-based variation operation with JS and KNN2W with relevant evaluation metrics as the most influential aspects.

Giant Photoluminescence Enhancement of Ga-Doped ZnO Microwires by X-Ray Irradiation.

Ga-doped zinc oxide (ZnO) microwires hold great promise for developing highly efficient light sources because of the wide bandgap with proper exciton binding energy. However, most microwires grown from one mainstream approach, i.e., chemical vapor deposition (CVD), are morphologically and crystallographically defective, exhibiting limited photoluminescence performances. Herein, a simple and effective X-ray irradiation strategy is demonstrated for enhancing the photoluminescence of Ga-doped ZnO microwire in ambient conditions. Under moderate doses (≤ 150Gy), the photoluminescence monotonically rockets up with X-ray dose increment and achieves nine-fold enhancement at a dose of ≈150Gy, recording high photoluminescence improvement of ZnO microwires to date. The elemental characteristics under different controlled irradiation atmospheres suggest the elimination of surface oxygen vacancy and the cross-section transmission electron microscope reveals prominent lattice relaxations after mild X-ray irradiation. In addition, the X-ray irradiated microwires further exhibit elevated electroluminescence by over three times. The enhanced photoluminescence and electroluminescence as well as long-term stability enable us to imagine the super-rapid applications of ZnO microwires in modern optoelectronic devices.

Point-of-interest recommendation based on the spatial-temporal graph

PurposePoint-of-interest (POI) recommendation techniques play a crucial role in mitigating information overload and delivering tailored services. To address limitations in conventional POI recommendation systems, constrained by sparse user-POI interactions and incomplete consideration of temporal dynamics, POI recommendation based on the spatial-temporal graph (STG-POI) is proposed.Design/methodology/approachSpatial-temporal sequence graphs from geographical locations and user interaction history data are constructed, which are used to mine spatial-temporal sequence information. Using the data filtered by the band-pass filter, graph neural networks with distance-awareness and sequence-awareness are applied to capture high-order spatial-temporal connections within diverse graph topologies. The model leverages contrastive learning for self-supervised disentanglement of graph representations, providing self-supervised signals for sequential and geographical intent perception, thereby achieving more precise POI personalization.FindingsCompared to the baseline model GSTN, experiments on the Foursquare and Gowalla data sets reveal that STG-POI improves testing AUC by 2.0%, 2.1%, 2.0% and decreases logloss by 1.9%, 3.3%, 0.3%, respectively. These results indicate the model’s effectiveness in capturing spatial-temporal information, surpassing mainstream POI recommendation baseline models.Originality/valueThis approach constructs a dual graph from user interaction data, harnessing sequential and geographical information as self-supervised signals. It yields decoupled representations of these influences, offering a comprehensive insight into user behaviors and preferences within location-based social networks, thus enhancing recommendation accuracy and interpretability. This approach addresses the challenge in graph convolutional network where only rough and smooth features are conducive to recommendation by using band-pass filters to significantly reduce computational complexity, thereby enhancing recommendation speed by filtering out noise data that does not contribute to recommendation performance. Experimental results indicate that this model surpasses current mainstream approaches in POI recommendation tasks, effectively integrating both geographical and temporal features.

Community-Engaged Development of Strengths-Based Nutrition Measures: The Indigenous Nourishment Scales.

Mainstream approaches to nutrition typically focus on diet consumption, overlooking multi-dimensional aspects of nutrition that are important to American Indian/Alaska Native (AI/AN) communities. To address health challenges faced by AI/AN communities, strengths-based measures of nutrition grounded in community worldviews are needed. In collaboration with AI/AN communities in Baltimore and Minneapolis, we developed the Indigenous Nourishment Scales through three phases. Phase 1 involved focus group discussions with nine community-research council (CRC) members (n = 2) and four in-depth interviews (n = 4) to gather perspectives on existing models of nutrition. Phase 2 refined scales through two additional focus group discussions (n = 2) with a total of nine participants and two in-depth interviews (n = 2). Finally, in Phase 3, we held ten (n = 10) cognitive interviews with AI/AN community members to refine the scales. Participants appreciated the measures' ability to provoke reflection on their relationship with nutrition and suggested adjustments to better capture cultural nuances, such as incorporating concepts like "being a good relative" to land. The Indigenous Nourishment Scales represent a departure from conventional approaches by encompassing multiple dimensions of nourishment, offering a framework that addresses epistemic injustices in nutrition measurement and grounds health measurement efforts directly in community perspectives and worldviews.

Compositional reservoir simulation with a high-resolution compact stencil adaptive implicit method

The adaptive implicit method (AIM) is the mainstream approach for compositional reservoir simulation. In its standard form, AIM uses single-point upwinding to reconstruct the fluxes across the interfaces of the control volume, and this leads to substantial numerical diffusion and loss of accuracy. Previous efforts to improve the accuracy of AIM focused on using high-order schemes to reconstruct the interfacial fluxes; those schemes introduced additional numerical nonlinearities to the system of equations or compromised the accuracy of the Jacobian by neglecting the high-order terms in its construction. In this work, we describe a high-resolution compact-stencil (HRCS) AIM. The new scheme is applied to compositional reservoir simulation. In addition to the mixed implicit/explicit time discretization of standard AIM, the HRCS AIM scheme uses a mixed time and space discretization. Specifically, we blend low- and high-order fluxes according to a well defined rule that uses a high-order reconstruction in the explicit regions of the domain and a low-order reconstruction in the implicit regions. This strategy ensures that additional nonlinearities introduced by the high-order reconstruction do not impact the Jacobian matrix, thus preserving the algebraic structure of standard AIM. The HRCS AIM method is demonstrated using several compositional problems. The results indicate substantial gains in accuracy with a small additional computational cost compared with standard AIM. Additionally, HRCS AIM is more robust and has a smaller computational cost compared with its full high-resolution counterpart.

Organised Violence, Inequality and Work: Violence Specialists as a Classed Occupation

Scholars have long debated what is the role of inequalities in organised violence, but the causal mechanisms remain unclear. I argue that mainstream approaches (e.g. deviance, subcultures, grievances, rational choice and Marxism) fall short because they overlook that organised violence is an intensive form of work. These perspectives often exoticise individuals involved in violence for profit or political purposes. Violence is not a deviant feature of social life but an integral part of collective action. By applying occupational lenses, I position those engaged in violence as a specialised class of manual workers recruited in the protection labour market. Scholars can acknowledge the occupational stratification observed before in research about organised violence by understanding violence specialists as workers. This essay explores how the sociology of work can enhance our understanding of inequality reproduction within criminal organisations, guerrillas, armies, police forces, mercenaries and private military companies.

Personal Identity Matters

Abstract This article considers whether process philosophy can provide a more promising basis for understanding the closely related problems of persons and personal identity than more traditional or mainstream philosophical approaches. In particular, the article focuses on whether process conceptions of persons and personal identity provide an approach that offers greater potential for resolving the duplication and fission examples within the literature in this area than the approaches taken by Bernard Williams and Derek Parfit.

Predictive pixel-wise optical encoding: towards single-shot high dynamic range moving object recognition

Conventional low dynamic range (LDR) imaging devices fail to preserve much information for further vision tasks because of the saturation effect. Thus, high dynamic range (HDR) imaging is an important imaging technology in extreme illuminance conditions, which enables a wide range of applications, including photography, autonomous driving, and robotics. Mainstream approaches require multi-shot methods because the conventional camera can only control the exposure globally. Although they perform well on static HDR imaging, they face a challenge with real-time HDR imaging for motional scenes because of the artifact and time latency caused by multi-shot and post-processing. To this end, we propose a framework, termed POE-VP, which achieves single-shot HDR imaging via a pixel-wise optical encoder driven by video prediction. We use highlighted motional license plate recognition as a downstream vision task to demonstrate the performance of POE-VP. From the results of simulation and real scene experiments, we validate that POE-VP outperforms conventional LDR cameras by more than 5 times in recognition accuracy and by more than 200% in information entropy. The dynamic range could reach 120 dB, and the captured data size is verified to be lower than mainstream multi-shot methods by 67%. The running time of POE-VP is also validated to satisfy the needs of high-speed HDR imaging.

Nesting landscape character and personality assessment to intensify rural landscape regionality and uniqueness

Urbanization impacts and homogenization pressures are threatening rural landscapes worldwide, leading to the loss of their regionality and uniqueness. Landscape Character Assessment (LCA) measures and identifies the uniqueness and distinctiveness of landscapes, making the unique value of rural resources easier to understand, protect, and perpetuate. Originating in the UK in the 1970s, LCA has become the mainstream approach for identifying landscape characteristics and is now a tool for managing landscape changes globally. Landscape Personality Assessment (LPA) offers another perspective for effectively measuring and analyzing landscape characteristics but has not been widely applied since its proposal. To distinguish between these two perspectives on landscape uniqueness identification, this paper conducted a literature review and comparison study, summarizing and comparing the characteristics of LCA and LPA in terms of evaluation scale, dimensions, indicators, and supported types of practices, further clarifying their respective strengths and weaknesses. To complement each other's perspectives on landscape characteristic identification, a framework nesting LCA and LPA was proposed to address the issue of the loss of rural landscape regionality in territorial spatial planning. This framework aims to couple evaluations of multi-level spatial characteristics and combine the identification of multi-temporal landscape characteristics. The goal is to enrich the theoretical and methodological system of rural landscape characteristic evaluation and provide more tools for the planning and conservation of the regionality of rural landscapes globally.

Grounded situation recognition under data scarcity

Grounded Situation Recognition (GSR) aims to generate structured image descriptions. For a given image, GSR needs to identify the key verb, the nouns corresponding to roles, and their bounding-box groundings. However, current GSR research demands numerous meticulously labeled images, which are labor-intensive and time-consuming, making it costly to expand detection categories. Our study enhances model accuracy in detecting and localizing under data scarcity, reducing dependency on large datasets and paving the way for broader detection capabilities. In this paper, we propose the Grounded Situation Recognition under Data Scarcity (GSRDS) model, which uses the CoFormer model as the baseline and optimizes three subtasks: image feature extraction, verb classification, and bounding-box localization, to better adapt to data-scarce scenarios. Specifically, we replace ResNet50 with EfficientNetV2-M for advanced image feature extraction. Additionally, we introduce the Transformer Combined with CLIP for Verb Classification (TCCV) module, utilizing features extracted by CLIP’s image encoder to enhance verb classification accuracy. Furthermore, we design the Multi-source Verb-Role Queries (Multi-VR Queries) and the Dual Parallel Decoders (DPD) modules to improve the accuracy of bounding-box localization. Through extensive comparative experiments and ablation studies, we demonstrate that our method achieves higher accuracy than mainstream approaches in data-scarce scenarios. Our code will be available at https://github.com/Zhou-maker-oss/GSRDS.

Precision Phenotyping in Crop Science: From Plant Traits to Gene Discovery for Climate‐Smart Agriculture

ABSTRACTThe global population is placing unprecedented demand on food systems, which can be met only through a complex interplay of technology, sustainable food production intensification methods and climate resilience. To address such compounded requirements, developing high‐yielding crop varieties using precise plant breeding methods bolstered with efficient and nondestructive plant trait documentation approaches is vital. High‐throughput crop phenotyping (HTCP) platforms have prominently emerged as a mainstream approach for reducing the phenotyping bottleneck in breeding programmes. HTCP has the potential to provide detailed quantitative information of large plant populations under different growth stages across diverse environmental regimes, facilitating accelerated plant breeding strategies. New imaging platforms also enable nondestructive characterization of a wide range of above and below‐ground crop parameters. The specificity in use of sensors, automation of data collection, large‐scale data handling systems and accurate analytical tools have a substantial role in dynamic crop monitoring and big data interpretation. HTCP platforms are capable of making precise measurements of a wide range of physiological, morphological, biochemical and stress responses in plants. Developments of sensors with improved precision, intervention of unmanned aerial vehicles, robotics, computed tomography and machine learning have given a dramatic developmental leap to precise and large‐scale crop phenotyping. This review provides an avenue for understanding various high‐throughput phenotyping platforms, working principles, current developments and contributions to high‐throughput phenotyping of various crops under laboratory and field conditions. A detailed comparative idea on the advantages and pitfalls of these available platforms can help researchers in choosing the right technology suiting specific practical requirements. Furthermore, the review aims to provide novel future prospects and developmental requirements that can potentially widen the application and utilization of these HTCP technologies in agriculture.

Identification of gas-liquid two-phase flow patterns based on flexible ultrasound array and machine learning

Ultrasound technology has been recognized as the mainstream approach for the identification of gas-liquid two-phase flow patterns, which holds great value in engineering domain. However, commercial rigid probes are bulky, limiting their adaptability to curved surfaces. Here, we propose a strategy for autonomous identification of flow patterns based on flexible ultrasound array and machine learning. The array features high-performance 1–3 piezoelectric composite material, stretchable serpentine wires, soft Eco-flex layers and a polydimethylsiloxane (PDMS) adhesive layer. The resulting ultrasound array exhibits excellent electromechanical characteristics and offers a large stretchability for an intimate interfacial contact to curved surface without the need of ultrasound coupling agents. We demonstrated that the flexible ultrasound array combined with machine learning can accurately identify gas-liquid two-phase flow patterns, in a circular pipeline. This work presents an effective tool for recognizing gas-liquid two-phase flow patterns, offering engineering opportunities in petroleum extraction and natural gas transportation.

Innovativeness of the European economies in the context of the modified European Innovation Scoreboard

Research background: Innovation is an important determinant of economic development, and its importance is growing with the advancement of digitization and the development of the Fourth Industrial Revolution. In turn, the assessment of the innovativeness of a country’s economy affects its investment attractiveness and international image. Purpose of this study: Given the factors presented above, the main aim of this study is to explore and compare the innovativeness of European economies in the context of the modified European Innovation Scoreboard. Methods: To achieve this goal, the authors propose two key modifications of the existing innovation assessment methodology contained in the European Innovation Scoreboard. The first concerns the principles of the normalization of assessment indicators and aims to eliminate extreme assessment values. The second proposes to remove the indicators which are difficult to measure, subjective, and not always useful: (1) those related to sustainable development (resource productivity; emissions of fine particulates; and environment-related technologies) and (2) those that take into account only small and medium-sized enterprises (SMEs introducing product innovations; SMEs introducing business process innovations; and innovative SMEs collaborating with others). Findings & value added: An alternative approach to assessing the innovativeness of the analyzed countries — as proposed by the authors — resulted in changes in the ranking of countries in relation to the European Innovation Scoreboard, but the changes were not significant. The major changes concerned Serbia (down four places), Belgium (down two places), Hungary (up three places), and Poland (up two places). In addition, five countries moved up one place in the ranking (the UK, the Netherlands, Luxembourg, Estonia, Slovenia, Spain, and Slovakia), and five lost one place in the ranking (Norway, Italy, Portugal, Greece, and Croatia). Thirteen countries had no changes. The obtained results allowed us to conclude that the countries with the best and the worst innovativeness maintained the assessment level regardless of changes in the methodology of its implementation. The objectification of rules had the strongest impact on the countries in the middle of the ranking. This study draws attention to the problem of the subjectivity of the innovation rankings. It has also been documented that modifying the selection of evaluation criteria may change the final position in the ranking, which, in the case of emerging and developing economies, may affect investor assessment and economic growth opportunities. Such considerations are important because they are outside the often uncritical mainstream approach to the assessment of the innovativeness of economies. The conclusions imply the need for a multisource analysis of innovation and a critical look at the proposed methodologies.

Research on Multi-Object Classification and Recognition Methods for Intelligent Vehicles Based on Deep Learning

With the development of intelligent transportation systems, multi-object classification and recognition for intelligent vehicles have become a key research direction in the fields of autonomous driving and intelligent transportation. Deep learning-based methods have become the mainstream approach to solving this problem. This paper reviews the deep learning-based multi-object classification and recognition techniques, analyzes the advantages and disadvantages of current mainstream methods, and discusses future research directions.

Bidirectional regulation of the cGAS-STING pathway in the immunosuppressive tumor microenvironment and its association with immunotherapy.

Adaptive anti-tumor immunity is currently dependent on the natural immune system of the body. The emergence of tumor immunotherapy has improved prognosis and prolonged the survival cycle of patients. Current mainstream immunotherapies, including immune checkpoint blockade, chimeric antigen receptor T-cell immunotherapy, and monoclonal antibody therapy, are linked to natural immunity. The cGAS-STING pathway is an important natural immunity signaling pathway that plays an important role in fighting against the invasion of foreign pathogens and maintaining the homeostasis of the organism. Increasing evidence suggests that the cGAS-STING pathway plays a key role in tumor immunity, and the combination of STING-related agonists can significantly enhance the efficacy of immunotherapy and reduce the emergence of immunotherapeutic resistance. However, the cGAS-STING pathway is a double-edged sword, and its activation can enhance anti-tumor immunity and immunosuppression. Immunosuppressive cells, including M2 macrophages, MDSC, and regulatory T cells, in the tumor microenvironment play a crucial role in tumor escape, thereby affecting the immunotherapy effect. The cGAS-STING signaling pathway can bi-directionally regulate this group of immunosuppressive cells, and targeting this pathway can affect the function of immunosuppressive cells, providing new ideas for immunotherapy. In this study, we summarize the activation pathway of the cGAS-STING pathway and its immunological function and elaborate on the key role of this pathway in immune escape mediated by the tumor immunosuppressive microenvironment. Finally, we summarize the mainstream immunotherapeutic approaches related to this pathway and explore ways to improve them, thereby providing guidelines for further clinical services.

The future of tourism: interview with Bernadett Papp on brining equity into debates on tourism development and destination governance

PurposeAs the Journal of Tourism Futures celebrates its 10th Anniversary, Dr Stefan Hartman (head of department at the European Tourism Futures Institute, NHL Stenden University of Applied Sciences) interviews Bernadett Papp (senior researcher at the European Tourism Futures Institute (ETFI) at NHL Stenden University of Applied Sciences) about her recent work in tourism practice and academic and future perspectives on topics relevant to the future of tourism.Design/methodology/approachA personal interview was held on September 16, 2024.FindingsBernadett Papp draws attention to the importance of bringing equity into debates on tourism development and destination governance. She sees a crucial role for governments, the redesign of institutional frameworks, monitoring and data collection and rethinking systems of governance.Originality/valueThe interview provides unique insights into the viewpoints of Papp, highlighting the importance of the still understudied topic of equitable tourism. A topic that still requires a lot of work before hitting mainstream approaches to destination development, even though it should have been on the policy agendas for many years already.

Hyperspectral image classification with token fusion on GPU

Hyperspectral images capture material nuances with spectral data, vital for remote sensing. Transformer has become a mainstream approach for tackling the challenges posed by high-dimensional hyperspectral data with complex structures. However, a major challenge they face when processing hyperspectral images is the presence of a large number of redundant tokens, which leads to a significant increase in computational load, adding to the model’s computational burden and affecting inference speed. Therefore, we propose a token fusion algorithm tailored to the operational characteristics of the hyperspectral image and pure transformer network, aimed at enhancing the final accuracy and throughput of the model. The token fusion algorithm introduces a token merging step between the attention mechanism and the multi-layer perceptron module in each Transformer layer. Experiments on four hyperspectral image datasets demonstrate that our token fusion algorithm can significantly improve inference speed without any training, while only causing a slight decrease in the pure transformer network’s classification accuracy.

Multiscale implicit frequency selective network for single‐image dehazing

AbstractImage dehazing is aimed to reconstruct a clear latent image from a degraded image affected by haze. Although vision transformers have achieved impressive success in various computer vision tasks, the limitations in scale and quality of available datasets have hindered the transformer effectiveness for image dehazing. Thus, convolutional neural networks (CNNs) remain the mainstream approach for image dehazing, offering robust performance and adaptability. We further explore the potential of CNNs in image dehazing by proposing a multiscale implicit frequency selection network (MIFSN). The proposed MIFSN enhances multiscale representation learning based on U‐shaped networks. As hazy and clear images considerably differ in high‐frequency components, we introduce an implicit frequency selection module to amplify high‐frequency components of features and generate candidate feature maps. Implicit frequency selection attention is then used to emphasize and merge beneficial frequency components. Results from extensive experiments on synthetic and real‐world datasets demonstrate the superior performance of MIFSN for image dehazing.

A hierarchical GNN across semantic and topological domains for predicting circRNA-microRNA interactions

Identifying circRNA-microRNA interactions (CMI) is a significant biomedical issue in recent years. This problem provides insights into using circRNA as biomarkers, developing cancer therapies and producing cancer vaccines. Using computational methods for identification is a more time-efficient and cost-effective approach. In computational methods, using graphs to represent and explore the CMI is a mainstream approach. However, existing relevant methods do not achieve optimal results by utilizing both the semantic information extracted from sequences and the topological information extracted from graph structures. To address this issue, we propose HGLMALLM, a graph contrastive learning method that learns node representation crossing both the semantic domain generated via motif-aware pre-trained LLMs and the topological domain extracted from hierarchical graph structures. Our method effectively addresses the issue in existing Message Passing Neural Network (MPNN) method that edge components losing heterogeneity after multiple iterations. Moreover, this method utilizes the heterogeneity of graph which is extended from the traditional bipartite graph to heterogeneous through the semantic domain. Two commonly used datasets were partitioned based on the distribution of node degrees. Then, we benchmarked our method against existing methods. In the independent testing set evaluation, it achieved a 3 % and 1 % improvement on two datasets. Our method demonstrated the best stability in ten-fold cross-validation on the training set. A test conducted on the peripheral components reveals robust performance of our model. A dataset collected from real scenarios was used to demonstrate the strong predictive ability of our method for identifying unidentified CMI.

Physics-informed ensemble learning with residual modeling for enhanced building energy prediction

Accurate modeling of building energy use is important to support a diverse spectrum of its downstream applications, such as building energy efficiency assessment, resilience analysis, smart control, etc. As mainstream approaches of building energy modeling, physics-based modeling builds on different fidelities of physics rules yet are usually compromised in modeling accuracy due to insufficiency of physics rules to capture real-world dynamics and incomplete input information. Data-driven approaches are computationally efficient, but black box (uninterpretable) in nature. For improved modeling of building energy use, this work proposes a physics-informed ensemble learning approach in building energy prediction through residual modeling. Specifically, we first analyze the components of building energy use data. Evidence suggests that the building energy use data can be decomposed into physics-driven part, occupant-driven part, and white noise. Second, high-fidelity physics-based building models (EnergyPlus) and low-fidelity ones (RC models) are developed to capture the physics-driven part while time series methods are explored as the residual modeling approach to capture the occupant-driven discrepancies between physics-based simulation and measured building energy use (i.e., residuals). Finally, the physics-informed ensemble learning is proposed to integrate physics-based and data-driven models for enhanced accuracy and robustness of building energy modeling. Results demonstrate 40–90% increase of accuracy between modeling and field observations compared to traditional physics-based modeling methods. Moreover, when the training dataset size is small, the proposed ensemble model overperforms the pure data-driven models, demonstrating its higher robustness in extrapolation scenarios. This work makes fundamental contributions to the development of convergent modeling approaches in the building modeling field.