Cross-Domain Social Rumor-Propagation Model Based on Transfer Learning.

Rumors in different topic domains have different text characteristics but similar emotional tendencies. To resolve the scarce-data problem in some rumor-topic domains, this study proposes a cross-domain rumor-propagation model, which is based on transfer learning. First, given the diversity and complexity of the rumor-propagation landscape, this study introduces a novel method, User-Retweet-Rumor2vec (URR2vec), which leverages the power of representation learning to uncover latent features within rumor topics. It also displays the forwarding relationship between users and rumors, user node information, and rumor-topic information in low-dimensional space. To capture the impact of human emotional cognition during rumor spreading, we also introduce a deep-learning model based on the natural language texts of rumor topics, which analyzes the sentiment in the text and uncovers the emotional correlations among users. Furthermore, a rumor-propagation prediction model based on the text-sentiment analysis-graph convolutional network (TSA-GCN) is proposed and pre-trained on existing rumor-topic data to ensure its prediction accuracy. Finally, considering the data sparsity at a rumor-topic outbreak, the trained propagation model is transferred to the rumor topic for prediction. Meanwhile, the rumor topic in different domains has different edges and conditional distribution, similar emotional characteristics, and network structure among the rumor topics. After fine-tuning the parameter and adding a domain adaptation layer in TSA-GCN, a domain adaptation model based on parameter and graph-structure migration is obtained.

Self-assembled lignin-based flame retardant hybrids carrying Cu2+ for poly(lactic acid) composites with improved fire safety and mechanical properties.

To enhance the flame retardancy and mechanical performance of PLA, a polyelectrolyte complex predicated on lignin was obtained by electrostatic mutual adsorption of ammonium polyphosphate (APP), polyethyleneimine (PEI), and copper ions as raw materials. The FT-IR spectra and EDX analysis confirmed the successful synthesis of a lignin-based flame retardant hybrid (APL-Cu2+) containing copper, phosphorus, and nitrogen elements. The combustion test results showed that the peak heat release rate and total heat release of the PLA composite containing 12 wt% APL-Cu2+ were decreased by 15.1 % and 18.2 %, respectively, as compared to those of pure PLA. The char residue morphology observation revealed that the addition of APL-Cu2+ could promote the formation of a highly dense and stable graphitized char layer, while TG-MS detected the emission of refractory gases such as ammonia gas, carbon dioxide, and water during combustion. The strong hydrogen bonding between APL-Cu2+ and the PLA matrix kept the composite maintaining good strength and toughness. The tensile strength and impact strength of PLA/6APL-Cu2+ increased by 4.73 % and 65.71 %, respectively, due to its high crystallinity and good interfacial compatibility. This work provides a feasible method to develop biobased flame retardant hybrids for PLA composites with better fire safety and improved mechanical properties.

Impacts of reclamation marsh restoration on greenhouse gas emission in the Sanjiang Plain, China.

To understand the variations in greenhouse gas fluxes during the process of returning cropland to wetland in the Sanjiang Plain, we selected naturally restored wetlands of 4, 7, 11, 16 and 20 years as research objects to compare with a cultivated site (soybean plantation for 13 years) and an uncultivated marsh dominated by Deyeuxia purpurea and Carex schmidtii. We measured carbon dioxide (CO2) and methane (CH4) fluxes using a static chamber-gas chromatography and explored the main influencing factors. The results showed that there were seasonal variations in growing-season CO2 and CH4 fluxes of the restored wetlands, with the seasonal trends in greenhouse gases becoming gradually similar to that of natural marsh with increasing restoration time. The mean growing-season CO2 fluxes increased during the early stage of restoration, but then decreased during the late stage, which decreased from 893.4 mg·m-2·h-1 to 494.0 mg·m-2·h-1 in the 4-year and 20-year sites, respectively. Mean CH4 fluxes increased with restoration time, ranging from a weak CH4 sink (soybean fields, -0.6 mg·m-2·h-1) to a CH4 source of 87.8 mg·m-2·h-1(20-year restored site). The CH4 fluxes of experimental plots were consistently lower than that of natural marsh (96.4 mg·m-2·h-1). Increases in water level and soil conductivity resulting from restoration were the main driving factors for the decrease in CO2 fluxes. The increases in water level and soil dissolved organic carbon resulting from restoration were the primary drivers for the increase of CH4 fluxes in the restored wetlands. The global warming potentials increased with restoration time, ranging from 27.8 t·CO2-eq·hm-2(soybean fields) to 130.8 t·CO2-eq·hm-2(plot of 20-year restoration), which gradually approached that of natural marsh (156.3 t·CO2-eq·hm-2). The emission of GHGs from restored wetlands in the Sanjiang Plain gradually approached those of natural marsh. Further monitoring is required to identify the maturity of restored wetlands for achieving greenhouse gas emissions equivalent to that of natural marshland.

Construction of highly ductile, UV-shielding polylactide/poly(butylene adipate-co-terephthalate) biocomposites with hyperbranched polysiloxane functionalized lignin as a biocompatibilizer.

Biodegradable polylactide/poly(butylene adipate-co-terephthalate) (PLA/PBAT) blends have been widely utilized as packaging materials. However, it is urgent to develop a biocompatibilizer to improve the interfacial interaction of the biodegradable immiscible polymer blends in practice. In this paper, a novel type of hyperbranched polysiloxane (HBPSi) with terminal methoxy groups was synthesized and then utilized to functionalize lignin through a hydrosilation reaction. The HBPSi modified lignin (lignin@HBPSi) was incorporated into immiscible PLA/PBAT blends to serve as a biocompatibilizer. The lignin@HBPSi was uniformly dispersed in the PLA/PBAT matrix with improved interfacial compatibility. Dynamic rheological results revealed that the addition of lignin@HBPSi reduced the complex viscosity, improving the processing ability of the PLA/PBAT composite. The PLA/PBAT composite containing 5 wt% lignin@HBPSi had a superior toughness with an elongation at break of 300.2 % and a slight enhancement in tensile stress (34.47 MPa). In addition, the presence of lignin@HBPSi contributed to blocking ultraviolet rays in the full ultraviolet band. This work provides a feasible way to develop highly ductile PLA/PBAT/lignin composites with good UV-shielding properties for the packaging applications.

Self-assembled biobased chitosan hybrid carrying N/P/B elements for polylactide with enhanced fire safety and mechanical properties.

The inherent shortcomings such as flammability, brittleness, and low crystallinity limit the broad applications of poly(lactic acid) (PLA). To improve the fire resistance and mechanical properties of PLA, a chitosan-based core-shell flame retardant additive (APBA@PA@CS) was prepared for PLA via the self-assembly of interionic interactions among chitosan (CS), phytic acid (PA), and 3-aminophenyl boronic acid (APBA). The peak heat release rate (pHRR) and total heat release rate (THR) of PLA composite containing 3 wt% APBA@PA@CS decreased from 460.1 kW/m2 and 75.8 MJ/m2 to 419.0 kW/m2 and 53.1 MJ/m2, respectively. The presence of APBA@PA@CS contributed to the formation of a high-quality char layer rich in phosphorus and boron in the condensed phase and released non-flammable gases in the gas phase to hinder the exchange of heat and O2, thereby having a synergistic flame retardant effect. Meanwhile, the tensile strength, elongation at break, impact strength, and crystallinity of PLA/APBA@PA@CS were increased by 3.7 %, 17.4 %, 5.3 %, and 55.2 %, respectively. This study provides a feasible route to construct a chitosan-based N/B/P tri-element hybrid to improve the fire safety performance and mechanical properties of PLA biocomposites.

Nanohybrid of Co3O4 Nanoparticles and Polyphosphazene-Decorated Ultra-Thin Boron Nitride Nanosheets for Simultaneous Enhancement in Fire Safety and Smoke Suppression of Thermoplastic Polyurethane.

Thermoplastic polyurethane (TPU) is widely used in daily life due to its characteristics of light weight, high impact strength, and compression resistance. However, TPU products are extremely flammable and will generate toxic fumes under fire attack, threatening human life and safety. In this article, a nanohybrid flame retardant was designed for the fire safety of TPU. Herein, Co3O4 was anchored on the surface of exfoliated ultra-thin boron nitride nanosheets (BNNO@Co3O4) via coprecipitation and subsequent calcination. Then, a polyphosphazene (PPZ) layer was coated onto BNNO@Co3O4 by high temperature polymerization to generate a nanohybrid flame retardant named BNNO@Co3O4@PPZ. The cone calorimeter results exhibited that the heat release and smoke production during TPU combustion were remarkably restrained after the incorporation of the nanohybrid flame retardant. Compared with pure TPU, the peak heat release rate (PHRR) decreased by 44.1%, the peak smoke production rate (PSPR) decreased by 51.2%, and the peak CO production rate (PCOPR) decreased by 72.5%. Based on the analysis of carbon residues after combustion, the significant improvement in fire resistance of TPU by BNNO@Co3O4@PPZ was attributed to the combination of quenching effect, catalytic carbonization effect, and barrier effect. In addition, the intrinsic mechanical properties of TPU were well maintained due to the existence of the PPZ organic layer.

A Context-aware Locality Measure for Inlier Pool Enrichment in Stepwise Image Registration.

We present a feature-based image registration method, the stepwise image registration (SIR), with a closed-form solution. Our SIR creates an inlier pool and a candidate pool as the initialization, and then gradually enriches the inlier pool and refines the transformation. In each step, the enriched correspondence exclusively tunes the transformation coefficient within the confirmed inlier pairs, instead of updating the mapping using the complete putative set. In turn, the refined transformation prunes inconsistent mismatches to alleviate the incoming matching ambiguity. The context-aware locality measure (CALM) is designed for dissimilarity measure. The capability of the CALM can be enhanced by the progressive inlier pool enrichment. Finally, a retrieval process is performed based on the finest CALM and alignment, by which the inlier pool is maximized. Extensive experiments of enrichment evaluation, feature matching, image registration, and image retrieval demonstrate the favorable performance of our SIR against state-of-the-art methods. The code and datasets are available at https://github.com/sucv/SIR.