Search

Domain information plays a crucial role in sentiment analysis. Neural networks that treat domain information as attention can further extract domain-related sentiment features from a shared feature pool, significantly enhancing the accuracy of sentiment analysis. However, when the sentiment polarity within the input text is inconsistent, these methods are unable to further model the relative importance of sentiment information. To address this issue, we propose a novel attention neural network that fully utilizes domain information while also accounting for the relative importance of sentiment information. In our approach, firstly, dual long short-term memory (LSTM) is used to extract features from the input text for domain and sentiment classification, respectively. Following this, a novel attention mechanism is introduced to fuse features to generate the attention distribution. Subsequently, the input text vector obtained based on the weighted summation is fed into the classification layer for sentiment classification. The empirical results from our experiments demonstrate that our method can achieve superior classification accuracies on Amazon multi-domain sentiment analysis datasets. [ABSTRACT FROM AUTHOR]

Optimizing the deployment of roadside units (RSUs) holds great potential for enhancing the delay performance of vehicular ad hoc networks. However, there has been limited focus on devising RSU deployment strategies tailored specifically for highway intersections. In this study, we introduce a novel probabilistic model to characterize events occurring around highway intersections. By leveraging this model, we analytically determine the expected event reporting delays for both highway segments and intersections. Subsequently, we propose an RSU deployment scheme specifically designed for highway intersections, aimed at minimizing the expected event reporting delay. To implement this scheme, we introduce an innovative algorithm named cooperative walking. Through illustrative examples, we demonstrate that our proposed RSU deployment strategy for highway intersections outperforms the commonly employed uniform RSU deployment scheme and the previously proposed balloon method in terms of delay performance. [ABSTRACT FROM AUTHOR]

To investigate the impact of frozen storage conditions on the physicochemical properties of soybean protein and explore the underlying mechanisms, this study focused on soybean isolate (SPI), ß-soybean companion globulin (7S), and soybean globulin (11S). The protein solutions were prepared at a concentration of 2% and subjected to freezing for 1 and 5 days. Subsequently, the protein content, physicochemical properties, secondary structure, sulfhydryl content, and chemical interaction forces were assessed and analyzed using UV spectrophotometry, Zeta potential measurements, SDS-PAGE, Fourier infrared spectroscopy, and endogenous fluorescence photoemission spectroscopy. The obtained results revealed that the solubility and total sulfhydryl content of SPI, 7S, and 11S exhibited a decreasing trend with prolonged freezing time. Among them, 11S demonstrated the largest decrease in solubility and total sulfhydryl content, followed by SPI, and 7S the least. During freezing, the aromatic amino acids of SPI, 7S, and 11S molecules were exposed, leading to increased hydrophobicity, protein aggregation, and particle size enlargement, and the structure of the protein changed from disordered structure to ordered structure. After freezing, the polarity of the microenvironment of SPI, 7S, and 11S increased, and their maximum fluorescence emission wavelengths were red-shifted. Notably, the largest red shift of SPI was from 332 nm to 335 nm. As freezing time increased, the contribution of hydrogen bonding increased, while the contribution of hydrophobic interactions decreased. This indicates that freezing affects the hydrophobic interactions, hydrogen bonding, and other chemical forces of the protein. The growth of ice crystals leads to the unfolding of protein molecular chains, exposure of internal hydrophobic groups, enhancement of hydrophobicity, and alters the secondary structure of the protein. [ABSTRACT FROM AUTHOR]

Background and Objectives: Pulsed electric field (PEF) treatment is currently a hot spot in food science research. The improvement and regulation of protein structure and physicochemical properties after PEF treatment are a new development in its processing technology. This study aimed to clarify the effects of moderate‐intensity PEF treatment on the structure and physicochemical properties of foxtail millet (Setaria italica) prolamin and its action rules. Findings: PEF treatment changed the spatial conformation of millet prolamin, increased its hydrophobicity and solubility in alcohol solution, but did not change the subunit composition distribution. It also promoted the increase in soluble aggregates, resulting in an increase in protein particle size. The α‐helix content of the protein decreased, the β‐sheet content increased, the number of disulfide bonds increased, and the protein structure became more orderly after treatment. Conclusions: The results showed that the structure and physicochemical properties of millet prolamin significantly changed after the moderate‐intensity PEF treatment. Significance and Novelty: The increased hydrophobicity of prolamin makes it a strong future candidate in medicinal excipients, food packaging materials, and biodegradable polymers. [ABSTRACT FROM AUTHOR]