Search

In order to solve the problems of the small capacity of structured data and uneven distribution among classes in machine learning tasks, a supervised generation method for structured data called WAGAN and a cyclic sampling method named SACS (Semi-supervised and Active-learning Cyclic Sampling), based on semi-supervised active learning, are proposed. The loss function and neural network structure are optimized, and the quantity and quality of the small sample set are enhanced. To enhance the reliability of generating pseudo-labels, a Semi-supervised Active learning Framework (SAF) is designed. This framework redistributes class labels to samples, which not only enhances the reliability of generated samples but also reduces the influence of noise and uncertainty on the generation of false labels. To mine the diversity information of generated samples, an uncertain sampling strategy based on spatial overlap is designed. This strategy incorporates the idea of spatial overlap and uses global and local sampling methods to calculate the information content of generated samples. Experimental results show that the proposed method performs better than other data enhancement methods on three different datasets. Compared to the original data, the average F 1 m a c r o value of the classification model is improved by 11.5%, 16.1%, and 19.6% relative to compared methods. [ABSTRACT FROM AUTHOR]

Green human resource management (GHRM) has become an indispensable strategy in green management. As the first step of GHRM, green recruitment and selection (GRS) has been emphasised as the important and probably the only practice to help the firm obtain green human capital directly. However, research has shown that not many firms implement GRS in practice, and there is a lack of studies exploring the barriers to GRS. This study fills this gap by examining the barriers to GRS across different industries in China. It adopted a qualitative and inductive approach, which is relatively rare in the GHRM field. Based on interviews with senior managers, a number of barriers were revealed in relation to the two components of GRS: paperless recruitment and green candidates. More specifically, two new factors that have not been reported in the existing literature were identified: the alignment between the job profile, green practices, and profit orientation. These two factors have both theoretical and practical implications. [ABSTRACT FROM AUTHOR]

As graph models become increasingly prevalent in the processing of scientific data, the exploration of effective methods for the mining of meaningful patterns from large-scale graphs has garnered significant research attention. This paper delves into the complexity of frequent subgraph mining and proposes a frequent subgraph mining (FSM) algorithm. This FSM algorithm is developed within a distributed graph iterative system, designed for the Big Cloud (BC) environment of the China Mobile Corp., and is based on the bulk synchronous parallel (BSP) model, named FSM-BC-BSP. Its aim is to address the challenge of mining frequent subgraphs within a single, large graph. This study advocates for the incorporation of a message sending and receiving mechanism to facilitate data sharing across various stages of the frequent subgraph mining algorithm. Additionally, it suggests employing a standard coded subgraph and sending it to the same node for global support calculation on the large graph. The adoption of the rightmost path expansion strategy in generating candidate subgraphs helps to mitigate the occurrence of redundant subgraphs. The use of standard coding ensures the unique identification of subgraphs, thus eliminating the need for isomorphism calculations. Support calculation is executed using the Minimum Image (MNI) measurement method, aligning with the downward closure attribute. The experimental results demonstrate the robust performance of the FSM-BC-BSP algorithm across diverse input datasets and parameter configurations. Notably, the algorithm exhibits exceptional efficacy, particularly in scenarios with low support requirements, showcasing its superior performance under such conditions. [ABSTRACT FROM AUTHOR]

Knowledge proficiency refers to the extent to which students master knowledge and reflects their cognitive status. To accurately assess knowledge proficiency, various pedagogical theories have emerged. Bloom's cognitive theory, proposed in 1956 as one of the classic theories, follows the cognitive progression from foundational to advanced levels, categorizing cognition into multiple tiers including "knowing", "understanding", and "application", thereby constructing a hierarchical cognitive structure. This theory is predominantly employed to frame the design of teaching objectives and guide the implementation of teaching activities. Additionally, due to the large number of students in real-world online education systems, the time required to calculate knowledge proficiency is significantly high and unacceptable. To ensure the applicability of this method in large-scale systems, there is a substantial demand for the design of a parallel prediction model to assess knowledge proficiency. The research in this paper is grounded in Bloom's Cognitive theory, and a Bloom Cognitive Diagnosis Parallel Model (BloomCDM) for calculating knowledge proficiency is designed based on this theory. The model is founded on the concept of matrix decomposition. In the theoretical modeling phase, hierarchical and inter-hierarchical assumptions are initially established, leading to the abstraction of the mathematical model. Subsequently, subject features are mapped onto the three-tier cognitive space of "knowing", "understanding", and "applying" to derive the posterior distribution of the target parameters. Upon determining the objective function of the model, both student and topic characteristic parameters are computed to ascertain students' knowledge proficiency. During the modeling process, in order to formalize the mathematical expressions of "understanding" and "application", the notions of "knowledge group" and "higher-order knowledge group" are introduced, along with a parallel method for identifying the structure of higher-order knowledge groups. Finally, the experiments in this paper validate that the model can accurately diagnose students' knowledge proficiency, affirming the scientific and meaningful integration of Bloom's cognitive hierarchy in knowledge proficiency assessment. [ABSTRACT FROM AUTHOR]

Learning patterns are crucial for predicting student dropout in educational settings, providing insights into students' behaviors and motivations. However, existing mainstream dropout prediction models have limitations in effectively mining these learning patterns and cannot mine these learning patterns in large-scale, distributed educational datasets. In this study, we analyze the representations of mainstream models and identify their inability to capture students' distinct learning patterns and personalized variations across courses. Addressing these challenges, our study adopts a federated learning approach, tailoring the analysis to leverage distributed data while maintaining privacy and decentralization. We introduce the Federated Learning Pattern Aware Dropout Prediction Model (FLPADPM), which utilizes a one-dimensional convolutional neural network (CNN) and a bidirectional long short-term memory (LSTM) layer within a federated learning framework. This model is designed to effectively capture nuanced learning patterns and adapt to variations across diverse educational settings. To evaluate the performance of LPADPM, we conduct an empirical evaluation using the KDD Cup 2015 and XuetangX datasets. Our results demonstrate that LPADPM outperforms state-of-the-art models in accurately predicting student dropout behavior. Furthermore, we visualize the representations generated by LPADPM, which confirm its ability to effectively mine learning patterns in different courses. Our results showcase the model's ability to capture and analyze learning patterns across various courses and institutions within a federated learning context. [ABSTRACT FROM AUTHOR]

Graph convolutional network (GCN) architecture is the basis of many neural networks and has been widely used in processing graph-structured data. When dealing with large and sparse data, deeper GCN models are often required. However, the models suffer from performance degradation as the number of layers increases. The mainstream attribution of the current research is over-smoothing, and there are also gradient vanishing, training difficulties, etc., so a consensus cannot be reached. In this paper, we theoretically analyze the degradation problem by adopting spectral graph theory to globally consider the propagation and transformation components of the GCN architecture, and conclude that the over-smoothing problem caused by the propagation matrices is not the key factor for performance degradation. Afterwards, in addition to using conventional experimental methods, we proposed an experimental analysis strategy under the guidance of random matrix theory to analyze the singular value distribution of the model weight matrix. We concluded that the key factor leading to the degradation of model performance is the transformation component. In the context of a lack of consensus on the problem of model performance degradation, the paper proposes a systematic analysis strategy, as well as theoretical and empirical evidence. [ABSTRACT FROM AUTHOR]

In order to find solutions to math word problems, some modules have been designed to check the generated expressions, but they neither take into account the symmetry between math word problems and their corresponding mathematical expressions, nor do they utilize the efficiency of pretrained language models in natural language understanding tasks. Anyway, designing fine-tuning tasks for pretrained language models that encourage cooperation with other modules to improve the performance of math word problem solvers is an unaddressed problem. To solve these problems, in this paper we propose a BERT-based model for solving math word problems with a supervision module. Based on pretrained language models, we present a fine-tuning task to predict the number of different operators in the expressions to learn the potential relationships between the problems and the expressions. Meanwhile, a supervision module is designed to check the incorrect expressions generated and improve the model's performance by optimizing the encoder. A series of experiments are conducted on three datasets, and the experimental results demonstrate the effectiveness of our model and its component's designs. [ABSTRACT FROM AUTHOR]

Location-based social networks (LBSNs) have collected massive geo-tagged information, enabling the derivation of user preference for point of interests (POIs) in support of personalized recommendation. The existing embedding techniques deal with multiple factors by embedding a separate model for each factor. As a result, the interaction amongst various factors cannot be captured properly. In addition, we notice that the effectiveness of personalized recommendation is closely related to the current time and location. It is obvious that users would check into a POI which fits their interests, even if the current location is far away from the POI or the time is inappropriate. Therefore, it is necessary to recommend the right POI according to the time and geographic location of the user. In other words, it is necessary to predict the most likely visiting event, including users, POI, event time, and event location. In this paper, we propose a probabilistic embedding model called Topic And Region Embedding (TARE), which embeds events by simulating the users' decision-making process. The results of TARE not only take various factors and their interaction into consideration but also consider the time and geographic location of events. Extensive experiments on three location-based social network datasets show that TARE achieves better performance in recommendation accuracy than existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Rice noodles are one of southern people's favorite foods in China, so it is important to find the suitable raw rice for rice noodle making. To study the effects of different rice varieties on the quality of fresh wet rice noodles and to explore the relationship between the quality of the rice and the quality of the fresh wet rice noodles, this study to compare the 12 hybrid rice varieties as raw materials analyzed the differences in the cooking quality, texture index, and sensory score of fresh wet rice noodles using the principal component analysis, membership function, and cluster analysis. The results showed that the quality of fresh wet rice noodles prepared from different hybrid rice materials differed significantly. The fresh wet rice noodles made from Liangyou 5836 are of good quality, and they are mainly characterized by a low rate of broken noodles and spit pulp value, high rice noodle hardness, good rice noodle elasticity, strong rice noodle chewiness, and low adhesiveness. Moreover, its sensory evaluation is also better than that of other varieties. The comprehensive evaluation of 12 hybrid rice varieties by subordinate function analysis also showed that Liangyou 5836 was the best. In addition, through principal component analysis and gray analysis, it was found that 14 related indicators of rice quality and fresh wet rice noodle quality were concentrated into four categories, among which gel consistency best reflects the quality of rice and fresh wet rice noodles. Through comprehensive analysis, it was found that an amylose content of about 22% and a gel consistency of less than 40 mm can be used as core indicators to screen varieties suitable for making rice noodles. This study is of great significance for the selection of hybrid rice for both rice quality and fresh wet rice noodle quality. [ABSTRACT FROM AUTHOR]

Accurate prediction of photovoltaic power is of great significance to the safe operation of power grids. In order to improve the prediction accuracy, a similar day clustering convolutional neural network (CNN)–informer model was proposed to predict the photovoltaic power. Based on correlation analysis, it was determined that global horizontal radiation was the meteorological factor that had the greatest impact on photovoltaic power, and the dataset was divided into four categories according to the correlation between meteorological factors and photovoltaic power fluctuation characteristics; then, a CNN was used to extract the feature information and trends of different subsets, and the features output by CNN were fused and input into the informer model. The informer model was used to establish the temporal feature relationship between historical data, and the final photovoltaic power generation power prediction result was obtained. The experimental results show that the proposed CNN–informer prediction method has high accuracy and stability in photovoltaic power generation prediction and outperforms other deep learning methods. [ABSTRACT FROM AUTHOR]

The fast evolution of anti-tumor agents embodies a deeper understanding of cancer pathogenesis. To date, chemotherapy, targeted therapy, and immunotherapy are three pillars of the paradigm for cancer treatment. The success of immune checkpoint inhibitors (ICIs) implies that reinstatement of immunity can efficiently control tumor growth, invasion, and metastasis. However, only a fraction of patients benefit from ICI therapy, which turns the spotlight on developing safe therapeutic strategies to overcome the problem of an unsatisfactory response. Molecular-targeted agents were designed to eliminate cancer cells with oncogenic mutations or transcriptional targets. Intriguingly, accumulating shreds of evidence demonstrate the immunostimulatory or immunosuppressive capacity of targeted agents. By virtue of the high attrition rate and cost of new immunotherapy exploration, drug repurposing may be a promising approach to discovering combination strategies to improve response to immunotherapy. Indeed, many clinical trials investigating the safety and efficacy of the combination of targeted agents and immunotherapy have been completed. Here, we review and discuss the effects of targeted anticancer agents on the tumor immune microenvironment and explore their potential repurposed usage in cancer immunotherapy. [ABSTRACT FROM AUTHOR]

Knowledge tracing, which is used to predict students' performance based on their previous practices, has attracted many researchers' attention. Especially in this rising period of intelligent education, many knowledge tracing methods have been developed. However, most of the existing knowledge tracing methods focus on the personality of practices and knowledge concepts but ignore the contexts related to the studying process. In this paper, we propose a context-based knowledge tracing model, which combines students' historical performance and their studying contexts during knowledge mastery. To be specific, we first define five studying contexts for performance prediction. The basic context is the current knowledge state of a student, which is described by their practice sequences. Then, a QR-matrix is defined to represent the relationship among questions, knowledge concepts, and responses, which describes the contexts of questions and knowledge. Furthermore, an improved LSTM model is proposed to capture the context of students' memory and forgetness, and a multi-head attention mechanism is designed to capture the context of students' behaviors. Finally, based on the captured contexts, the prediction model ContextKT is established. Our prediction model is evaluated on two real educational datasets. The experimental results show our model is effective and efficient in student performance prediction, and it outperforms the other existing methods. [ABSTRACT FROM AUTHOR]

Improving the conductivity of metal particle inks is a hot topic of scientific research. In this paper, a method for preparing metal-filled particles was proposed. By adding filled particles to the ink, the size distribution of particles could be changed to form a bimodal distribution structure in accordance with Horsfield's stacking model. The filling particles had small volume and good fluidity, which could fill the gaps between the particles after printing and improve its electrical conductivity without significantly changing the metal solid content in the ink. Experimental results show that the silver content of the ink slightly increased from 15 wt% to 16.5 wt% after adding filled particles. However, the conductivity of the ink was significantly improved, and after sintering, the resistivity of the ink decreased from 70.2 μΩ∙cm to 31.2 μΩ∙cm. In addition, the filling particles prepared by this method is simple and has a high material utilization rate, which could be applied to the preparation of other kinds of metal particle inks. [ABSTRACT FROM AUTHOR]

Urban-intensive areas are responsible for an estimated 80% of greenhouse gas emissions, particularly carbon dioxide. The urban–rural fringe areas emit more greenhouse gases than urban centers. The purpose of this study is to analyze the spatial pattern and driving factors of carbon emissions in urban–rural fringe mixed-use communities, and to develop planning methods to reduce carbon emissions in communities. This study identifies mixed-use communities in East Asian urban–rural fringe areas as industrial, commercial, tourism, and rental-apartment communities, subsequently using the emission factor method to calculate carbon emissions. The statistical information grid analysis and geographic information systems spatial analysis method are employed to analyze the spatial pattern of carbon emission and explore the relationship between established space, industrial economy, material consumption, social behavior, and carbon emission distribution characteristics by partial least squares regression, ultimately summing up the spatial pattern of carbon emission in the urban–rural fringe areas of East Asia. Results show that (1) mixed-use communities in the East Asian urban–rural fringe areas face tremendous pressure to reduce emissions. Mixed-use community carbon emissions in the late urbanization period are lower than those the early urbanization. (2) Mixed-use community carbon emission is featured by characteristics, such as planning structure decisiveness, road directionality, infrastructure directionality, and industrial linkage. (3) Industrial communities produce the highest carbon emissions, followed by rental-apartment communities, business communities, and tourism communities. (4) The driving factor that most affects the spatial distribution of carbon emissions is the material energy consumption. The fuel consumption per unit of land is the largest driver of carbon emissions. Using the obtained spatial pattern and its driving factors of carbon emissions, this study provides suggestions for planning and construction, industrial development, material consumption, and convenient life guidance. [ABSTRACT FROM AUTHOR]