Search

Building accurate human behavior models is necessary for ambient intelligence. However, human activity recognition (HAR) in continuously monitored physical space meets many challenges to achieve a good performance when using only simple computing resources. In this work, we model HAR as an edge classification problem for a collaborative event graph of context entities in a sequential bipartite graph form. We design a semantic learning framework, called KGAR, to perform HAR by mining, encoding, and exploiting deep semantic knowledge of activities in an end-to-end fashion. KGAR has three components: preprocessor, KGEncoder, and predictor. The preprocessor builds offline a tiny knowledge graph of activities, to model and capture multidimensional semantic relationships between activities and core context entities. KGEncoder encodes the knowledge graph of activities using improved graph neural networks (GNNs) models, to handle confusing context patterns. The predictor may be deployed using some lightweight deep neural network to produce real-time labels. Experimental results show that using KGEncoder in KGAR improves the performance of original deep neural networks by 25% - 439% on five datasets. The time of labeling each sensor event during testing with event streams is less than 0.5 ms. We have also conducted extensive experimental study to show that KGAR outperforms different types of models in more complex activity scenarios. We believe KGAR can be used for real-time HAR in real life with its high prediction performance and a low computing requirement.

Heterogeneous palladium catalysts with high efficiency, high Pd atom utilization, simplified separation, and recycle have attracted considerable attention in the field of synthetic chemistry. Herein, we reported a zirconium-based two-dimensional metal–organic framework (2D-MOF)-based Pd(II) photocatalyst (Zr–Ir–Pd) by merging the Ir photosensitizers and Pd(II) species into the skeletons of the 2D-MOF for the Pd(II)-catalyzed oxidation reaction. Morphological and structural characterization identified that Zr–Ir–Pd with a specific nanoflower-like structure consists of ultrathin 2D-MOF nanosheets (3.85 nm). Due to its excellent visible-light response and absorption capability, faster transfer and separation of photogenerated carriers, more accessible Pd active sites, and low mass transfer resistance, Zr–Ir–Pd exhibited boosted photocatalytic activity in catalyzing sterically hindered isocyanide insertion of diarylalkynes for the construction of fused tetracyclic heterocycles, with up to 12 times the Pd catalyst turnover number than the existing catalytic systems. In addition, Zr–Ir–Pd inhibited the competitive agglomeration of Pd(0) species and could be reused at least five times, owing to the stabilization of 2D-MOF on the single-site Pd and Ir sites. Finally, a possible mechanism of the photocatalytic synthesis of fused tetracyclic heterocycles catalyzed by Zr–Ir–Pd was proposed.

The article advances a 3-D adaptive regularization nonlinear inversion of the logging while drilling (LWD) ultradeep multicomponent resistivity (LWD-UDMCR) by the Gauss-Newton (GN) method. We manage to reconstruct the pixel-based horizontal and vertical conductivities simultaneously in a goal domain outside an arbitrary dipping borehole. The piecewise constant functions are used to describe the spatial distribution of the block-based and the pixel-based conductivity. The background formation is assumed as the horizontally layered transversely isotropic (TI) media, and the background electromagnetic (EM) fields are determined analytically by the transmission line method (TLM). We then use the 3-D finite volume method (FVM) of secondary field coupled potentials and parallel direct sparse solver (PARDISO) to simulate the tool responses and Fréchet derivatives simultaneously. Through the projection operator and OpenMP parallel technique, we further enhance the computational efficiency of the pixel-based explicit Fréchet derivatives and set up a complete normalization linearized response. After that, the large normal equation from the quadratic objective function is solved by the preconditioned conjugate gradient (PCG) to determine the gradient of the objective function. By properly controlling the maximum component of the gradient per iteration step, we acquire an adaptive regularization factor so that the stabilization of the inversion solution is assured as well as the realization of the best fit of the input data with the modeling logs. Finally, numerical tests validate the algorithm and antinoise ability.

Heterogeneous Pd(ii) catalysts with high efficiency, high Pd atom utilization, high stability and easy recyclability are always a pursuit in the field of green organic synthesis chemistry. Herein, a two-dimensional (2D) metal–organic framework (MOF) based Pd(ii) catalyst Zr-BTB-Bpy-Pd is synthesized through continuous three-step coordination assembly strategy. A variety of structural characterization techniques establish Zr-BTB-Bpy-Pd as a nanoflower structure with ultrathin 2D nanoplate petals (3.9 nm), which provides fully exposed single-site Pd(ii) active sites and a similar homogeneous reaction space. Consequently, Zr-BTB-Bpy-Pd exhibits excellent activity for the cyclization reactions of isonitriles with N-acyl-o-alkynylanilines, far exceeding the three-dimensional (3D) MOF based Pd(ii) catalyst Pd-UiO-67-Bpy and homogeneous Pd(OAc)2, with turnover frequency (TOF) values being 2.3 and 864 times higher, respectively. The Zr-BTB-Bpy-Pd catalyst could be used for the gram-scale synthesis of 2-aminoquinoline derivatives in up to 90% yield with a TOF of 1900 and in five consecutive reaction cycles without significant loss of activity.

The comprehensive utilization of the phosphorus–potassium associated ore has become significant to the new energy industry in China. This study proposes a new method for utilization of the phosphorus–potassium associated ore by phosphoric acid leaching. The mineral phase of the phosphorus–potassium associated ore was analyzed by X-ray diffraction and mineral liberation analysis. The leaching characteristics, mechanism, and kinetics of phosphorus, potassium, aluminum, magnesium, and iron were systematically investigated. Phosphorus and potassium leaching processes were controlled by the internal diffusion of the shrinking core model with apparent activation energies of 30.43 and 23.81 kJ/mol. The optimized leaching process conditions were as follows: stirring speed of 400 rpm, particle size of 150–300 μm, phosphorus concentration (P2O5) of 14%, phosphorus-to-calcium ratio of 9.65, reaction temperature of 40 °C, and reaction time of 60 min. The phosphorus leaching rate was 96.18 ± 0.92%, and the potassium leaching rate was 2.60 ± 0.08%. The acid-insoluble residue was mainly composed of potassium feldspar, indicating that effective separation of phosphorus and potassium was achieved through phosphoric acid leaching.

• MHPD is proposed to evaluate the spreading influence of nodes on hypergraphs. • The most influential nodes are selected by the MHPD-greedy and MHPD-heuristic. • The time cost and performance of the proposed methods are theoretically analyzed. • The proposed methods significantly outperform baselines on accuracy and efficiency. • Visualization and statistics are utilized to verify the effectiveness of the methods. Influence maximization problem (IM) has been extensively applied in fields such as viral marketing, rumor control, and infectious disease prevention. However, research on the IM problem has primarily focused on ordinary networks, with limited attention devoted to hypergraphs. Firstly, we propose an efficient evaluation method, i.e., the multiple-hop probability dissemination method (MHPD), aiming to accurately and rapidly evaluate the propagation capacity of selected nodes. The MHPD method is a universal approach that is applicable to various network types, including hypergraphs, and it accommodates multiple probabilistic spreading models. Then, based on MHPD, we propose two novel algorithms for solving IM problem, i.e., MHPD-greedy and MHPD-heuristic. MHPD-greedy employs MHPD to evaluate the marginal benefits of nodes and iteratively adds nodes with the maximum marginal benefit to the seed set. MHPD-heuristic utilizes MHPD to evaluate the propagation capacity of each node and select the top- K nodes as the seeds. Experimental results on eight real-world hypergraphs and eight synthetic hypergraphs demonstrate that MHPD is capable of achieving near-identical accuracy in evaluating the propagation capability of both individual nodes and seed sets, while only incurring an average time overhead of merely 0.25 % compared to Monte Carlo method. In comparison with seven cutting-edge algorithms, MHPD-heuristic demonstrates superior solution accuracy. Notably, MHPD-greedy maintains 98.81 % of the solution accuracy while requiring only 0.11 % of the time cost compared to the Greedy method. Furthermore, MHPD-greedy achieves an average performance improvement of 23.4 % over the best baseline algorithm. [ABSTRACT FROM AUTHOR]

Rapid global changes since the last deglaciation can be well documented in marginal sea sediments, while isolating individual paleoenvironmental signals is still challenging. Here, we identified magnetic minerals and unmixed their variations in sediments from the northwestern South China Sea to decipher environmental variations since the last deglaciation. The variation in the hematite to goethite ratio indicates a relatively drier mid‐Holocene and a wetter early and late Holocene in the Red River catchment. The position shift and intensity variation of Western Pacific Subtropical High may account for the Holocene precipitation changes. A higher chemical weathering intensity accompanied by a relatively more fine‐grained magnetite input during the last deglaciation suggests decoupling of the Asian Summer Monsoon and chemical weathering in the catchment, which might have been caused by the shelf exposure during the deglaciation. In addition, the relative abundance of biogenic magnetite increased with the sea level and possible deep‐water temperature rise. Therefore, magnetic minerals such as hematite, goethite, detrital and biogenic magnetite are markedly potential fingerprints for continent‐ocean environmental changes. The global climate, which has changed rapidly since the last deglaciation, has been recorded by marginal sea sediments. However, the mixing of environmental factors renders it difficult to isolate individual information. Here, we unmixed the magnetic mineralogical signals of sediments from the northwestern South China Sea since the last deglaciation. We found that a higher hematite to goethite ratio (Hm/Gt) in our core represents higher precipitation in Red River catchment. The Hm/Gt variation indicates a relatively drier mid‐Holocene and a wetter early and late‐Holocene in the Red River catchment, which is consistent with the precipitation records in Southern China but opposite to that of Northern China. The chemical weathering of the Red River catchment was decoupled from the Asian Summer Monsoon, exhibiting a higher weathering extent (more fine‐grained magnetite was generated) during the last deglaciation due to secondary weathering on the continental shelf. Moreover, the relative abundance of magnetofossils in our core was consistent with changes of the sea level and deep‐water environments. Therefore, the isolation of magnetic mineral information from marginal sea sediments can help to reconstruct the continent‐ocean environmental changes. A drier mid‐Holocene and wetter late‐Holocene occurred in Red River catchment controlled by the shift of Western Pacific Subtropical HighA decoupling between chemical weathering in Red River catchment and summer monsoon was caused by shelf exposure during the deglaciationThe relative content of magnetofossils in northwestern South China Sea is controlled by the sea level and deep‐water environment changes A drier mid‐Holocene and wetter late‐Holocene occurred in Red River catchment controlled by the shift of Western Pacific Subtropical High A decoupling between chemical weathering in Red River catchment and summer monsoon was caused by shelf exposure during the deglaciation The relative content of magnetofossils in northwestern South China Sea is controlled by the sea level and deep‐water environment changes

Abutment pressure distribution is different when a longwall panel is passing through the abandoned gate roads in a damaged coal seam. According to the geological condition of panel E13103 in Cuijiazhai Coal Mine in China, theoretical analysis and finite element numerical simulation were used to determine the front pressure distribution characteristics when the longwall face is 70, 50, 30, 20, 10, and 5 m from the abandoned roadways. The research results show that the influence range of abutment pressure is 40 to 45 m outby the face, and the peak value of front abutment pressure is related to the distance between the face and abandoned roadways. When the distance between the longwall face and abandoned roadways is reduced from 50 to 10 m, the front abutment pressure peak value kept increasing. When the distance is 10 m, it has reached the maximum. The peak value is located in 5 to 6 m outby the faceline. When the distance between the longwall face and abandoned roadways is reduced from10 to5 m, the front abutment pressure sharply decreases, the intact coal yields and is even in plastic state. The peak value transfers to the other side of the abandoned roadways. The research results provide a theoretical basis for determining the advance support distance of two roadways in the panel and the reinforcement for face stability when the longwall face is passing through the abandoned roadways.

Widespread fake news on social media threatens public security and the cyber environment, making fake news detection an essential area of study. The majority of existing fake news detection methods rely on news content (e.g., text and images) and/or social contexts (e.g., comment interactions between posts) to determine the veracity of news. However, existing methods still have the following drawbacks: (1) Overreliance on sufficient reliable labeled data. (2) Lack of robustness to noise and fraudster-designed harmful disguises. (3) Inability to differentiate between the multiple intentions behind retweet and comment behaviors, resulting in generating entangled representations. To address the above aforementioned three issues, we introduce contrastive learning and disentangled representation learning for fake news detection. Specifically, to mine supervised signals from unlabeled data and improve the model's robustness, we design a hierarchical contrastive learning framework that includes multiple data augmentation strategies and three contrastive learning tasks. In addition, to infer the latent intentions of retweets and comments between posts, we propose the disentangled graph encoder (Disen-GraphEnc) and disentangled sequence encoder (Disen-SeqEnc). Extensive experiments demonstrate the superiority of our model over other state-of-the-art methods and is resistant to limited training data and noise attacks. Our code is available on the GitHub (https://github.com/senllh/DHCF). [ABSTRACT FROM AUTHOR]

The anhysteretic remanent magnetization (ARM) is an important magnetic parameter for quantifying the concentration of “fine”, mostly sub‐micron ferrimagnetic particles in rocks and sediments. A sound understanding of ARM carriers is needed to interpret magnetic and environmental information in sedimentary archives. This is often not achieved for marine sediments, which routinely contain several magnetic mineral components. Here, we analyze marine sediment records over different timescales for four sediment cores from the Eastern Pacific Ocean (Hole 1218A), the Antarctic margin (core P4‐1), the Arctic Ocean (core ARC5‐ICE4), and the South China Sea (core L07), using coercivity spectra analyses of ARM, first‐order reversal curve principal component analysis (FORC‐PCA), transmission electron microscopy (TEM), and unmixing of isothermal remanent magnetization (IRM) curves. By combining rock magnetic and TEM results, we find that the ARM is mainly carried by non‐interacting single domain (SD) biogenic magnetite (magnetofossils) in the studied sediments over all timescales. Low‐coercivity magnetic particles (detrital and fine‐grained extracellular magnetite) also contribute to the bulk ARM. Variable magnetofossil chain structures are detected and have a significant influence on ARM data interpretation. Quantitative FORC‐PCA endmember analyses based on quantile contours and coercivity distributions provide valuable information on the nature of endmembers and improve ARM data interpretation. We provide an integrated strategy to help reduce interpretational ambiguities related to ARM in future studies. The anhysteretic remanent magnetization (ARM) is an important parameter that is used to assess information about Earth's magnetic field variations and environmental changes. To interpret ARM signals robustly, we must understand the sedimentary magnetic minerals that control ARM changes. However, the origin of ARM changes in marine sediments is not well understood. Here, we analyze marine sediment records from the Eastern Pacific Ocean, the Antarctic margin, the Arctic Ocean, and the South China Sea to identify the carriers of ARM in the respective sediments. Our results indicate that all of the sediments contain abundant remains of magnetotactic bacteria (known as magnetofossils). Magnetofossil abundance variations dominate ARM on both 10,000 to million‐year timescales. Part of the bulk ARM is also contributed by other low‐coercivity magnetite particles. Furthermore, ARM is influenced significantly by magnetofossil chain structure variations. Our results have important implications for future studies involving use of ARM in magnetofossil‐rich sediments. Integrated component analyses are presented for four sediment cores to identify magnetic particle sourcesMagnetofossils are the predominant anhysteretic remanent magnetization (ARM) carrier with detrital and extracellular magnetite also contributing to the ARMQuantitative analyses provide valuable insights into the nature of components to improve ARM data interpretation of marine sediments Integrated component analyses are presented for four sediment cores to identify magnetic particle sources Magnetofossils are the predominant anhysteretic remanent magnetization (ARM) carrier with detrital and extracellular magnetite also contributing to the ARM Quantitative analyses provide valuable insights into the nature of components to improve ARM data interpretation of marine sediments

The relationships among the abundance of magnetofossils, the ensuing magnetic properties, and the controlling paleoenvironmental factors in marine sediments remain broadly unexplored. Here, we identify magnetofossils in core XB1 from the northwestern South China Sea (SCS) since end of the Last Glacial. Using rock magnetic and electron microscopic data, we propose a model that links the anisotropy of magnetic susceptibility fabric and the abundance of magnetofossils. The magnetofossil concentration in sediments increases significantly during the 14.7–4.7 ka period, which in turns leads to inverse magnetic fabrics and near‐horizontal of the minimum magnetic susceptibility axes. Further, we show that the abundance of magnetofossils is linked to paleoenvironmental changes in the northwestern SCS. The production and preservation of magnetofossils during the 14.7–4.7 ka period are promoted by an intensified East Asian summer monsoon and sluggish deep‐water ventilation, while the paucity of magnetofossils after 4.7 ka is attributed to high oxygen content. The anisotropy of magnetic susceptibility (AMS) is a useful magnetic property of rocks for a wide range of scientific questions. Typically, the sedimentary AMS fabric carried by coarser‐grained magnetic particles (i.e., multi‐domain, MD) exhibits an oblate shape with a horizontal maximum magnetic susceptibility axis. However, single‐domain (SD) magnetic particles (e.g., nano‐sized magnetofossil magnetites) complicate the AMS fabric, through their intrinsic inverse magnetic fabric with the minimum magnetic susceptibility axis aligned along the horizontal plane. For natural samples, quantification of the contributions of SD and MD particles to AMS is essential for accurate interpretation of the AMS. This paper puts forward a novel approach that links the AMS and the abundance of magnetofossils. By applying this new method to sediments from the South China Sea, we found that the inverse magnetic fabrics in the studied sediments are associated with high magnetofossils concentration, which can be affected by various factors including the East Asian monsoon, nutrient supply or deep‐water ventilation. Values of the minimum susceptibility axis inclination are associated with the abundance of magnetofossilsThe magnetofossils can be quantified by the proportion of inverse component in sedimentsThe strong deep‐water ventilation hinders the production and preservation of magnetofossils in the South China Sea Values of the minimum susceptibility axis inclination are associated with the abundance of magnetofossils The magnetofossils can be quantified by the proportion of inverse component in sediments The strong deep‐water ventilation hinders the production and preservation of magnetofossils in the South China Sea

Supported metal nanocatalyst is one of the efficient tools for organic transformations. However, catalyst deactivation caused by the migration, aggregation, and leaching of active metal species in the reaction process remains challenging. Herein, a metal–organic framework (MOF), MIL-101, was employed to covalently graft the PPh3ligand on its surface and then supported palladium nanoparticles (Pd NPs), affording Pd/MIL-101-PPh3. A variety of spectral characterizations and DFT calculation reveal that there is an electron-donating effect of the MOF surface PPh3toward Pd NPs, which markedly boosts the activation of the carbon-halogen bond in aryl halides. Consequently, Pd/MIL-101-PPh3exhibits excellent activity for the three-component reaction of 2-iodoaniline, CO2, and isocyanide, as well as Suzuki–Miyaura and Heck coupling reactions, far exceeding amino-functionalized Pd/MIL-101-NH2, naked Pd/MIL-101, and other commercial-supported Pd catalysts. Furthermore, Pd/MIL-101-PPh3can also frustrate the migration, aggregation, and leaching of reactive Pd species in the reaction process due to the molecular fence effect generated by MOF surface functionalization.