Search

A great many practical applications have observed knowledge evolution, i.e., continuous born of new knowledge, with its formation influenced by the structure of historical knowledge. This observation gives rise to evolving knowledge graphs whose structure temporally grows over time. However, both the modal characterization and the algorithmic implementation of evolving knowledge graphs remain unexplored. To this end, we propose EvolveKG–a general framework that enables algorithms in the static knowledge graphs to learn the evolving ones. EvolveKG quantifies the influence of a historical fact on a current one, called the effectiveness of the fact, and makes knowledge prediction by leveraging all the cross-time knowledge interaction. The novelty of EvolveKG lies in Derivative Graph–a weighted snapshot of evolution at a certain time. Particularly, each weight quantifies knowledge effectiveness through a temporarily decaying function of consistency and attenuation, two proposed factors depicting whether or not the effectiveness of a fact fades away with time. Besides, considering both knowledge creation and loss, we obtain higher prediction accuracy when the effectiveness of all the facts increases with time or remains unchanged. Under four real datasets, the superiority of EvolveKG is confirmed in prediction accuracy. [ABSTRACT FROM AUTHOR]

As a carrier of knowledge, papers have been a popular choice since ancient times for documenting everything from major historical events to breakthroughs in science and technology. With the booming development of science and technology, the number of papers has been growing exponentially. Just like the fact that Internet of Things (IoT) allows the world to be connected in a flatter way, how will the network formed by massive academic papers look like? Most existing visualization methods can only handle up to hundreds of thousands of node size, which is much smaller than that of academic networks which are usually composed of millions or even more nodes. In this paper, we are thus motivated to break this scale limit and design a new visualization method particularly for super-large-scale academic networks (VSAN). Nodes can represent papers or authors while the edges means the relation (e.g., citation, coauthorship) between them. In order to comprehensively improve the visualization effect, three levels of optimization are taken into account in the whole design of VSAN in a progressive manner, i.e., bearing scale, loading speed, and effect of layout details. Our main contributions are two folded: 1) We design an equivalent segmentation layout method that goes beyond the limit encountered by state-of-the-arts, thus ensuring the possibility of visually revealing the correlations of larger-scale academic entities. 2) We further propose a hierarchical slice loading approach that enables users to observe the visualized graphs of the academic network at both macroscopic and microscopic levels, with the ability to quickly zoom between different levels. In addition, we propose a "jumping between nebula graphs" method that connects the static pages of many academic graphs and helps users to form a more systematic and comprehensive understanding of various academic networks. Applying our methods to three academic paper citation datasets in the AceMap database confirms the visualization scalability of VSAN in the sense that it can visualize academic networks with more than 4 million nodes. The super-large-scale visualization not only allows a galaxy-like scholarly picture unfolding that were never discovered previously, but also returns some interesting observations that may drive extra attention from scientists. [ABSTRACT FROM AUTHOR]

Active canopy sensors (ACSs) are great tools for diagnosing crop nitrogen (N) status and grain yield prediction to support precision N management strategies. Different commercial ACSs are available and their performances in crop N status diagnosis and recommendation may vary. The objective of this study was to determine the potential to minimize the differences of two commonly used ACSs (GreenSeeker and Crop Circle ACS-430) in maize (Zea mays L.) N status diagnosis and recommendation with multi-source data fusion and machine learning. The regression model was based on simple regression or machine learning regression including ancillary information of soil properties, weather conditions, and crop management information. Results of simple regression models indicated that Crop Circle ACS-430 with red-edge based vegetation indices performed better than GreenSeeker in estimating N nutrition index (NNI) (R2 = 0.63 vs. 0.50–0.51) and predicting grain yield (R2 = 0.56–0.57 vs. 0.49). The random forest regression (RFR) models using vegetation indices and ancillary data greatly improved the prediction of NNI (R2 = 0.81–0.82) and grain yield (R2 = 0.87–0.89), regardless of the sensor type or the vegetation index used. Using RFR models, moderate degree of accuracy in N status diagnosis was achieved based on either GreenSeeker or Crop Circle ACS-430. In comparison, using simple regression models based on spectral data only, the accuracy was significantly lower. When these two ACSs were used independently, they performed similarly in N fertilizer recommendation (R2 = 0.57–0.60). Hybrid RFR models were established using vegetation indices from both ACSs and ancillary data, which could be used to diagnose maize N status (moderate accuracy) and make side-dress N recommendations (R2 = 0.62–0.67) using any of the two ACSs. It is concluded that the use of multi-source data fusion with machine learning model could improve the accuracy of ACS-based N status diagnosis and recommendation and minimize the performance differences of different active sensors. The results of this research indicated the potential to develop machine learning models using multi-sensor and multi-source data fusion for more universal applications. [ABSTRACT FROM AUTHOR]

Geoscience knowledge graph (GKG) can organize various geoscience knowledge into a machine understandable and computable semantic network and is an effective way to organize geoscience knowledge and provide knowledge-related services. As a result, it has gained significant attention and become a frontier in geoscience. Geoscience knowledge is derived from many disciplines and has complex spatiotemporal features and relationships of multiple scales, granularities, and dimensions. Therefore, establishing a GKG representation model conforming to the characteristics of geoscience knowledge is the basis and premise for the construction and application of GKG. However, existing knowledge graph representation models leverage fixed tuples that are limited in fully representing complex spatiotemporal features and relationships. To address this issue, this paper first systematically analyzes the categorization and spatiotemporal features and relationships of geoscience knowledge. On this basis, an adaptive representation model for GKG is proposed by considering the complex spatiotemporal features and relationships. Under the constraint of a unified spatiotemporal ontology, this model adopts different tuples to adaptively represent different types of geoscience knowledge according to their spatiotemporal correlation. This model can efficiently represent geoscience knowledge, thereby avoiding the isolation of the spatiotemporal feature representation and improving the accuracy and efficiency of geoscience knowledge retrieval. It can further enable the alignment, transformation, computation, and reasoning of spatiotemporal information through a spatiotemporal ontology. [ABSTRACT FROM AUTHOR]

Rapid methods allowing for non-destructive crop monitoring are imperative for accurate in-season nitrogen (N) status assessment and precision N management. The objectives of this paper were to (1) compare the performance of a leaf fluorescence sensor Dualex 4 and an active canopy reflectance sensor Crop Circle ACS-430 for estimating maize (Zea mays L.) N status indicators across growth stages; (2) evaluate the potential of N status prediction across growth stages using the reflectance parameters acquired from the canopy sensor at an early growth stage; and, (3) investigate the prospect of combining the active canopy sensor and leaf fluorescence sensor data to estimate N nutrition index (NNI) indirectly using a general model across growth stages. The results indicated that data from both sensors were closely related to NNI across stages. However, using the direct NNI estimation method, among the tested indices, only the N balance index (NBI) could diagnose N status satisfactorily, based on the Kappa statistics. The effect of growth stages on proximal sensing was reduced by incorporating the information of days after sowing. It was found that the leaf fluorescence sensor performed relatively better in estimating plant N concentration whereas the canopy reflectance sensor performed better in aboveground biomass estimation. Their combination significantly improved the reliability of N diagnosis, including NNI prediction. In addition, the study confirmed that N status can be assessed by predicting aboveground biomass at the later stages using the canopy reflectance measurements at an early stage. Furthermore, the integrated NBI was verified to be a more robust and sensitive N status indicator than the chlorophyll concentration index. It is concluded that combining active canopy sensor data, of an early growth stage (e.g. V8), with leaf fluorescence sensor data, modified using days after sowing, can improve the accuracy of corn N status diagnosis across growth stages. [ABSTRACT FROM AUTHOR]

In this paper, a CO2 laser induced discharge plasma extreme ultraviolet (EUV) source experimental device was established. The optical emission spectroscopy was used to diagnose the characteristics of the plasma, and the evolution of electron temperature and electron density with time was obtained. The influence of discharge voltage on plasma parameters was analyzed and discussed. The EUV radiation characteristics of the plasma were investigated by self-made grazing incidence EUV spectrometer. The EUV radiation intensity and conversion efficiency were discussed. [ABSTRACT FROM AUTHOR]

Foam targets are expected to be more efficient candidates than solid targets for laser produced plasma (LPP) for extreme ultraviolet (EUV) and X-ray radiation sources due to the expected plasma conditions that can be optimized regarding plasma opacities, volumetrics heating effects, and the produced ions debris characteristics. In this paper, a comparison of ion dynamics between low-density foam and solid Ni plasma was systematically investigated at CMUXE. The foam Ni target (density 0.6 g/cm3) and solid Ni target (density 8.9 g/cm3) were irradiated with 1064 nm Nd:YAG laser in vacuum. A Faraday cup (FC) was used to record the ion flux and time-of-flight (TOF) signals. A lower and wider TOF signal was observed for foam Ni plasma on the time scale. The average ion energy and peak of the TOF signal of solid Ni plasma were much higher than that of the foam Ni plasma. However, the total charge values between foam and solid Ni plasma were comparable indicating a more volumetric absorption of laser energy for foam Ni. The average ion energy and peak of the TOF signal of solid Ni showed a stronger angular and laser energy dependence than that of foam Ni. The plume shape of the solid Ni plasma appeared as an oblong ellipse at each time, while that of foam Ni plasma tended to be more circular, especially at early times. The results of mass ablation rate were consistent with the FC signals and showed a more intense plasma shielding for solid Ni. [ABSTRACT FROM AUTHOR]

Since the beginning of the 21st century, the geoscience research has been entering a significant transitional period with the establishment of a new knowledge system as the core and with the drive of big data as the means. It is a revolutionary leap in the research of geoscience knowledge discovery from the traditional encyclopedic discipline knowledge system to the computer-understandable and operable knowledge graph. Based on adopting the graph pattern of general knowledge representation, the geoscience knowledge graph expands the unique spatiotemporal features to the Geoscience knowledge, and integrates geoscience knowledge elements, such as map, text, and number, to establish an all-domain geoscience knowledge representation model. A federated, crowd intelligence-based collaborative method of constructing the geoscience knowledge graph is developed here, which realizes the construction of high-quality professional knowledge graph in collaboration with global geo-scientists. We also develop a method for constructing a dynamic knowledge graph of multi-modal geoscience data based on in-depth text analysis, which extracts geoscience knowledge from massive geoscience literature to construct the latest and most complete dynamic geoscience knowledge graph. A comprehensive and systematic geoscience knowledge graph can not only deepen the existing geoscience big data analysis, but also advance the construction of the high-precision geological time scale driven by big data, the compilation of intelligent maps driven by rules and data, and the geoscience knowledge evolution and reasoning analysis, among others. It will further expand the new directions of geoscience research driven by both data and knowledge, break new ground where geoscience, information science, and data science converge, realize the original innovation of the geoscience research and achieve major theoretical breakthroughs in the spatiotemporal big data research. [ABSTRACT FROM AUTHOR]

Electronic devices require the printed circuit board (PCB) to support the whole structure, but the assembly of PCBs suffers from welding problem of the electronic components such as surface mounted devices (SMDs) resistors. The automated optical inspection (AOI) machine, widely used in industrial production, can take the image of PCBs and examine the welding issue. However, the AOI machine could commit false negative errors and dedicated technicians have to be employed to pick out those misjudged PCBs. This paper proposes a machine learning based method to improve the accuracy of AOI. In particular, we propose an adjacent pixel RGB value based method to pre-process the image from the AOI machine and build a customized deep learning model to classify the image. We present a practical scheme including two machine learning procedures to mitigate AOI errors. We conduct experiments with the real dataset from a production line for three months, the experimental results show that our method can reduce the rate of misjudgment from 0.3%–0.5% to 0.02%–0.03%, which is meaningful for thousands of PCBs each containing thousands of electronic components in practice. [ABSTRACT FROM AUTHOR]

The rapid growth of online social networks (OSNs) has ultimately facilitated information spreading and changed the economics of mobile networks. It is important to understand how to spread information as widely as possible. In this paper, we aim to seek powerful information initial spreaders with an efficient manner. We use the mean-field theory to characterize the process of information spreading based on the Susceptible Infected (SI) model and validate that the prevalence of information depends on the network density. Inspired by this result, we seek the initial spreaders from closely integrated groups of nodes, i.e., dense groups (DGs). In OSNs, DGs distribute dispersedly over the network, so our approach can be fulfilled in a distributed way by seeking the spreaders in each DG. We first design a DG Generating Algorithm to detect DGs, where nodes within the DG have more internal connections than external ones. Second, based on the detected DGs, we design a criterion to seek powerful initial spreaders from each DG. We conduct experiments as well as statistical analysis on real OSNs. The results show that our approach provides a satisfactory performance as well as computational efficiency. [ABSTRACT FROM AUTHOR]

A high performance plasmonic sensor based on a metal-insulator-metal (MIM) waveguide coupled with a double-cavity structure consisting of a side-coupled rectangular cavity and a disk cavity is proposed. The transmission characteristics of the rectangular cavity and disk cavity are analyzed theoretically and the improvements of performance for the double-cavity structure compared with a single cavity are studied. The influence of structural parameters on the transmission spectra and sensing performance are investigated in detail. A sensitivity of 1136 nm/RIU with a high figure of merit of 51,275 can be achieved at the resonant wavelength of 1148.5 nm. Due to the high performance and easy fabrication, the proposed structure may be applied in integrated optical circuits and on-chip nanosensors. [ABSTRACT FROM AUTHOR]

When a glass substrate was irradiated by three different temporal shapes of laser sources, namely, linetime-shape laser, triangle-time-shape laser, and parabolatime-shape laser, the mathematical models were proposed, and the temperature distribution and the resulting thermal stress were calculated by the finite-element-method (FEM) software ANSYS. With these three types of lasers having the same output laser energy, the resulting thermal stress induced in the glass substrate was analyzed. The results showed that, with the same output laser energy, the thermal stress produced in glass heated by line-time-shape laser is higher than that produced in glass heated by the other two shapes of lasers. [ABSTRACT FROM AUTHOR]

Recent advances in fine-tuning large-scale vision-language pre-trained models (VL-PTMs) have shown promising results in quick adaption to downstream tasks. However, prior research often lacks comprehensive investigation into out-of-distribution (OOD) generalization. Fine-tuning has a potential risk of overfitting, especially on few-shot OOD datasets when significant distribution shifts occur between the few-shot training examples and test sets. Previous research on fine-tuning’s robustness to distribution shifts does not consider different characteristics of distribution shifts and may not effectively handle noisy data with spurious correlations. To address these challenges, we propose the Vision-Language Alignment Learning under Affinity and Divergence Principles (VLAD) to adapt VL-PTMs to robust few-shot OOD generalization with theoretical guarantees. Built upon the large-scale pre-trained vision-language foundation model CLIP, we leverage frozen language embeddings as invariant anchors to protect against distribution shifts, while using adapter layers to fine-tune pre-trained visual features for improved vision-language alignment. Besides, we introduce affinity and divergence principles to further mitigate overfitting during the vision-language aligning process by increasing class discrimination and suppressing non-causal features. More importantly, we offer theoretical evidence highlighting the superiority of general language knowledge in achieving more robust OOD generalization performance. The tighter upper bound of the OOD generalization errors by the proposed regularization loss is also shown in theoretical analysis. Our approach is substantiated by extensive experiments and ablation studies on diverse datasets, validating our theoretical findings. The code is available at https://github.com/LinLLLL/VLAD. [ABSTRACT FROM AUTHOR]