Search

In this study, a series of novel nanofibers based on gelatin (GA) loading with purple potato anthocyanin (PPA) and syringic acid (SA) were obtained by electrospinning technology. The effects of SA on mechanical properties, thermal stability, antioxidant capacity, and antimicrobial activity of the GA/PPA nanofibers were systematically characterized. The scanning electron microscopy observation results revealed a smooth surface on the nanofibers. The incorporation of SA enhanced the viscosity of the electrospun solutions, and it increased the average diameter of nanofibers from 0.17 μm to 0.28 μm. The tensile strength and thermal stability of the obtained nanofibers were enhanced with the addition of a suitable level of SA (1.5%, w/v), which strengthened the intermolecular interaction. The GA/PPA/SA nanofibers presented over 80% antioxidant capacity and strong antibacterial activity against E. coli and S. aureus. Meanwhile, the sensitivity responses of nanofibers to NH3 revealed that GA/PPA/SA II nanofibers (1.5% w/v SA) presented good sensitivity of colorimetric behavior to ammonia. A pork spoilage test was performed to evaluate practical application of the nanofibers, and an obvious color change (dark purple to green) was observed. These results indicated GA/PPA/SA II nanofibers can be utilized as an active and intelligent multipurpose packaging material to preserve and track the freshness of pork. [ABSTRACT FROM AUTHOR]

Trimethylamine-N-oxide (TMAO), a diet-derived cometabolite linked to cardiometabolic disease, has been associated with elevated dietary status, particularly in people with kidney failure and adults with dietary modulations. However, the influence of the current diet on TMAO levels in free-living children has not been adequately described. This study was to explore associations of food compositions and dietary diversity with urinary TMAO and its precursor concentrations. Urinary TMAO and its precursor concentrations of 474 healthy children from the Sheyang Mini Birth Cohort were quantified by ultra-performance liquid chromatography–Q Exactive high-resolution mass spectrometer (UPLC-Q Exactive HRMS). Individual food compositions from 24 h dietary recall data were classified into 20 groups and diversity scores were calculated according to the guidelines of the Food and Agriculture Organization of the United Nations (FAO). Associations of urinary TMAO and its precursors with food compositions and dietary diversity scores were assessed by generalized linear regression models. In models adjusted for potential confounders, urinary TMAO was significantly associated with intakes of fish (β, regression coefficient = 0.155, p < 0.05) and vegetables (β = 0.120, p < 0.05). Eggs intake showed positive associations with TMAO's precursors (trimethylamine: β = 0.179, p < 0.05; choline: β = 0.181, p < 0.05). No association between meat intake and TMAO was observed, whereas meat and poultry intakes were related to the levels of acetyl-L-carnitine and L-carnitine (β: 0.134 to 0.293, p < 0.05). The indicators of dietary diversity were positively correlated to TMAO concentration (β: 0.027 to 0.091, p < 0.05). In this free-living children-based study, dietary factors were related to urinary TMAO and its precursors, especially fish, meat, and eggs. As such, dietary diversity was positively related to the level of TMAO. [ABSTRACT FROM AUTHOR]

Currently, most text style transfer methods encode the text into a style-independent latent representation and decode it into new sentences with the target style. Due to the limitation of the latent representation, previous works can hardly get satisfactory target style sentence especially in terms of semantic remaining of the original sentence. We propose a "Mask and Generation" structure, which can obtain an explicit representation of the content of original sentence and generate the target sentence with a transformer. This explicit representation is a masked text that masks the words with the strong style attribute in the sentence. Therefore, it can preserve most of the semantic meaning of the original sentence. In addition, as it is the input of the generator, it also simplified this process compared to the current work who generate the target sentence from scratch. As the explicit representation is readable and the model has better interpretability, we can clearly know which words changed and why the words changed. We evaluate our model on two review datasets with quantitative, qualitative, and human evaluations. The experimental results show that our model generally outperform other methods in terms of transfer accuracy and content preservation. [ABSTRACT FROM AUTHOR]

While there have been significant advances in detecting emotions in text, in the field of utterance-level emotion recognition (ULER), there are still many problems to be solved. In this paper, we address some challenges in ULER in dialog systems. (1) The same utterance can deliver different emotions when it is in different contexts. (2) Long-range contextual information is hard to effectively capture. (3) Unlike the traditional text classification problem, for most datasets of this task, they contain inadequate conversations or speech. (4) To better model the emotional interaction between speakers, speaker information is necessary. To address the problems of (1) and (2), we propose a hierarchical transformer framework (apart from the description of other studies, the "transformer" in this paper usually refers to the encoder part of the transformer) with a lower-level transformer to model the word-level input and an upper-level transformer to capture the context of utterance-level embeddings. For problem (3), we use bidirectional encoder representations from transformers (BERT), a pretrained language model, as the lower-level transformer, which is equivalent to introducing external data into the model and solves the problem of data shortage to some extent. For problem (4), we add speaker embeddings to the model for the first time, which enables our model to capture the interaction between speakers. Experiments on three dialog emotion datasets, Friends, EmotionPush, and EmoryNLP, demonstrate that our proposed hierarchical transformer network models obtain competitive results compared with the state-of-the-art methods in terms of the macro-averaged F1-score (macro-F1). [ABSTRACT FROM AUTHOR]

Featured Application: The model proposed in this paper can be deployed to the enterprise gateway, dynamically monitor network activities, and connect with the firewall to protect the company network from attacks. It can be deployed in cloud computing environment or in software-defined networks to monitor network behavior and alerts in real time. Machine learning plays an important role in building intrusion detection systems. However, with the increase of data capacity and data dimension, the ability of shallow machine learning is becoming more limited. In this paper, we propose a fuzzy aggregation approach using the modified density peak clustering algorithm (MDPCA) and deep belief networks (DBNs). To reduce the size of the training set and the imbalance of the samples, MDPCA is used to divide the training set into several subsets with similar sets of attributes. Each subset is used to train its own sub-DBNs classifier. These sub-DBN classifiers can learn and explore high-level abstract features, automatically reduce data dimensions, and perform classification well. According to the nearest neighbor criterion, the fuzzy membership weights of each test sample in each sub-DBNs classifier are calculated. The output of all sub-DBNs classifiers is aggregated based on fuzzy membership weights. Experimental results on the NSL-KDD and UNSW-NB15 datasets show that our proposed model has higher overall accuracy, recall, precision and F1-score than other well-known classification methods. Furthermore, the proposed model achieves better performance in terms of accuracy, detection rate and false positive rate compared to the state-of-the-art intrusion detection methods. [ABSTRACT FROM AUTHOR]

Konjac glucomannan (KGM) hydrogel has favorable gel-forming abilities, but its insufficient swelling capacity and poor control release characteristics limit its application. Therefore, in this study, oxidized hyaluronic acid (OHA) was used to improve the properties of KGM hydrogel. The influence of OHA on the structure and properties of KGM hydrogels was evaluated. The results show that the swelling capacity and rheological properties of the composite hydrogels increased with OHA concentration, which might be attributed to the hydrogen bond between the KGM and OHA, resulting in a compact three-dimensional gel network structure. Furthermore, epigallocatechin gallate (EGCG) was efficiently loaded into the KGM/OHA composite hydrogels and liberated in a sustained pattern. The cumulative EGCG release rate of the KGM/OHA hydrogels was enhanced by the increasing addition of OHA. The results show that the release rate of composite hydrogel can be controlled by the content of OHA. These results suggest that OHA has the potential to improve the properties and control release characteristics of KGM hydrogels. [ABSTRACT FROM AUTHOR]

Phishing has become one of the biggest and most effective cyber threats, causing hundreds of millions of dollars in losses and millions of data breaches every year. Currently, anti-phishing techniques require experts to extract phishing sites features and use third-party services to detect phishing sites. These techniques have some limitations, one of which is that extracting phishing features requires expertise and is time-consuming. Second, the use of third-party services delays the detection of phishing sites. Hence, this paper proposes an integrated phishing website detection method based on convolutional neural networks (CNN) and random forest (RF). The method can predict the legitimacy of URLs without accessing the web content or using third-party services. The proposed technique uses character embedding techniques to convert URLs into fixed-size matrices, extract features at different levels using CNN models, classify multi-level features using multiple RF classifiers, and, finally, output prediction results using a winner-take-all approach. On our dataset, a 99.35% accuracy rate was achieved using the proposed model. An accuracy rate of 99.26% was achieved on the benchmark data, much higher than that of the existing extreme model. [ABSTRACT FROM AUTHOR]

Gestational exposure to manganese (Mn), an essential trace element, is associated with fetal and childhood physical growth. However, it is unclear which period of growth is more significantly affected by prenatal Mn exposure. The current study was conducted to assess the associations of umbilical cord-blood Mn levels with birth outcomes and childhood continuous physical development. The umbilical cord-blood Mn concentrations of 1179 mother–infant pairs in the Sheyang mini birth cohort were measured by graphite furnace atomic absorption spectrometry (GFAAS). The association of cord-blood Mn concentrations with birth outcomes, and the BMI z-score at 1, 2, 3, 6, 7 and 8 years old, were estimated separately using generalized linear models. The relationship between prenatal Mn exposure and BMI z-score trajectory was assessed with generalized estimating equation models. The median of cord-blood Mn concentration was 29.25 μg/L. Significantly positive associations were observed between Mn exposure and ponderal index (β, regression coefficient = 0.065, 95% CI, confidence interval: 0.021, 0.109; p = 0.004). Mn exposure was negatively associated with the BMI z-score of children aged 1, 2, and 3 years (β = −0.383 to −0.249, p < 0.05), while no significant relationships were found between Mn exposure and the BMI z-score of children at the age of 6, 7, and 8 years. Prenatal Mn exposure was related to the childhood BMI z-score trajectory (β = −0.218, 95% CI: −0.416, −0.021; p = 0.030). These results indicated that prenatal Mn exposure was positively related to the ponderal index (PI), and negatively related to physical growth in childhood, which seemed most significant at an early stage. [ABSTRACT FROM AUTHOR]

Cross-domain relation extraction has become an essential approach when target domain lacking labeled data. Most existing works adapted relation extraction models from the source domain to target domain through aligning sequential features, but failed to transfer non-local and non-sequential features such as word co-occurrence which are also critical for cross-domain relation extraction. To address this issue, in this paper, we propose a novel tripartite graph architecture to adapt non-local features when there is no labeled data in the target domain. The graph uses domain words as nodes to model the co-occurrence relation between domain-specific words and domain-independent words. Through graph convolutions on the tripartite graph, the information of domain-specific words is propagated so that the word representation can be fine-tuned to align domain-specific features. In addition, unlike the traditional graph structure, the weights of edges innovatively combine fixed weight and dynamic weight, to capture the global non-local features and avoid introducing noise to word representation. Experiments on three domains of ACE2005 datasets show that our method outperforms the state-of-the-art models by a big margin. [ABSTRACT FROM AUTHOR]

Intrusion detection systems play an important role in preventing security threats and protecting networks from attacks. However, with the emergence of unknown attacks and imbalanced samples, traditional machine learning methods suffer from lower detection rates and higher false positive rates. We propose a novel intrusion detection model that combines an improved conditional variational AutoEncoder (ICVAE) with a deep neural network (DNN), namely ICVAE-DNN. ICVAE is used to learn and explore potential sparse representations between network data features and classes. The trained ICVAE decoder generates new attack samples according to the specified intrusion categories to balance the training data and increase the diversity of training samples, thereby improving the detection rate of the imbalanced attacks. The trained ICVAE encoder is not only used to automatically reduce data dimension, but also to initialize the weight of DNN hidden layers, so that DNN can easily achieve global optimization through back propagation and fine tuning. The NSL-KDD and UNSW-NB15 datasets are used to evaluate the performance of the ICVAE-DNN. The ICVAE-DNN is superior to the three well-known oversampling methods in data augmentation. Moreover, the ICVAE-DNN outperforms six well-known models in detection performance, and is more effective in detecting minority attacks and unknown attacks. In addition, the ICVAE-DNN also shows better overall accuracy, detection rate and false positive rate than the nine state-of-the-art intrusion detection methods. [ABSTRACT FROM AUTHOR]

Three-dimensional nanofibers cryogels (NFCs) with both thermally-tolerant and mechanically-robust properties have potential for wide application in biomedical or food areas; however, creating such NFCs has proven to be extremely challenging. In this study, konjac glucomannan (KGM)/poly (lactic acid) (PLA)-based novel NFCs were prepared by the incorporation of the mussel-inspired protein polydopamine (PDA) via a facile and environmentally-friendly electrospinning and freeze-shaping technique. The obtained KGM/PLA/PDA (KPP) NFCs were characterized by field emission scanning electron microscopy (FE-SEM), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), thermogravimetric analysis (TGA), differential scanning calorimetry (DSC) and compressive and tensile test. The results showed that the hierarchical cellular structure and physicochemical properties of KPP NFCs were dependent on the incorporation of PDA content. Moreover, the strong intermolecular hydrogen bond interactions among KGM, PLA and PDA also gave KPP NFCs high thermostability and mechanically-robust properties. Thus, this study developed a simple approach to fabricate multifunctional NFCs with significant potential for biomedical or food application., Competing Interests: The authors declare no conflict of interest.

Mitochondria are essential organelles and important targets for environmental pollutants. The detection of mitochondrial biogenesis and generation of reactive oxygen species (ROS) and p53 levels following low-dose methylmercury (MeHg) exposure could expand our understanding of underlying mechanisms. Here, the sensitivity of immortalized human neural progenitor cells (ihNPCs) upon exposure to MeHg was investigated. We found that MeHg altered cell viability and the number of 5-ethynyl-2'-deoxyuridine (EdU)-positive cells. We also observed that low-dose MeHg exposure increased the mRNA expression of cell cycle regulators. We observed that MeHg induced ROS production in a dose-dependent manner. In addition, mRNA levels of peroxisome-proliferator-activated receptor gammacoactivator-1α (PGC-1α), mitochondrial transcription factor A (TFAM) and p53-controlled ribonucleotide reductase (p53R2) were significantly elevated, which were correlated with the increase of mitochondrial DNA (mtDNA) copy number at a concentration as low as 10 nM. Moreover, we examined the expression of microRNAs (miRNAs) known as regulatory miRNAs of p53 (i.e., miR-30d, miR-1285, miR-25). We found that the expression of these miRNAs was significantly downregulated upon MeHg treatment. Furthermore, the overexpression of miR-25 resulted in significantly reducted p53 protein levels and decreased mRNA expression of genes involved in mitochondrial biogenesis regulation. Taken together, these results demonstrated that MeHg could induce developmental neurotoxicity in ihNPCs through altering mitochondrial functions and the expression of miRNA., Competing Interests: The authors declare no conflict of interest.

To elucidate the structure-antioxidant activity relationships of chitosan gallate (CG), a series of CG derivatives with different degrees of substitution (DS's) and molecular weights (MWs) were synthesized from chitosan (CS) and gallic acid (GA) via a free radical graft reaction. A higher MW led to a lower DS of CG. The structures of CG were characterized by FT-IR and ¹H NMR, and results showed that GA was mainly conjugated to the C-2 and C-6 positions of the CS chain. The antioxidant activity (the DPPH radical scavenging activity and reducing power) were enhanced with an increased DS and a decreased MW of CG. A correlation between antioxidant activities and the DS and MW of CG was also established. In addition, a suitable concentration (0~250 μg/mL) of CG with different MWs (32.78~489.32 kDa) and DS's (0~92.89 mg·GAE/g CG) has no cytotoxicity. These results should provide a guideline to the application of CG derivatives in food or pharmacology industries.

The ninety day toxicity and toxicokinetics of fluorochloridone (FLC) were accessed in Wistar rats. Animals were gavaged with FLC at doses of 31.25 mg/kg, 125 mg/kg and 500 mg/kg for ninety days, followed by thirty days for recovery. On the 1st, 60th, 75th and 90th days of the dosing phase, plasma of ten animals of all groups treated with FLC was collected for toxicokinetic analysis of FLC by an UPLC-MS/MS method. Numerous changes in body weight, hematology, serum chemistry, and organ weight ratios were observed by the 45th and 90th dosing day. Most changes in groups treated with FLC were absent on the last recovery day. Testis and epididymis lesions were consistently seen in histopathological observations on the 45th, 90th dosing day and the last recovery day. Repeated administration of FLC increased the level of testosterone in serum in male rats on the 90th dosing day. FLC plasma concentrations could be detected in all animal drug-treated groups during the dosing phase, and a dose proportional relationship was seen between FLC dose and AUC or Cmax. This study will support future studies on the mechanism of FLC-induced toxicity.