Sentiment classification is a key task in sentiment analysis, reviews mining, and other text mining applications. Various models have been proposed to build sentiment classifiers, but the classification performances of some existing methods are not good enough. Meanwhile, as a subproblem of sentiment classification, positive and unlabeled learning (PU learning) problem widely exists in real-world cases, but it has not been given enough attention. In this article, we aim to solve the two problems in one framework. We first build a model for traditional sentiment classification based on adversarial learning, attention mechanism, and long short-term memory (LSTM) network. We further propose an enhanced adversarial learning method to tackle PU learning problem. We conducted extensive experiments in three real-world datasets. The experimental results demonstrate that our models outperform the compared methods in both traditional sentiment classification problem and PU learning problem. Furthermore, we study the effect of our models on word embedding. Finally, we report and discuss the sensitivity of our models to parameters.
Erigeron Canadensis L. (E. canadensis) is a widely distributed invasive weed species in China. Potentially anticancer qualities may exist in its essential oils (EOs). The purpose of this study was to analyze the components of the EOs of E. canadensis and their effects on the normal liver cell lines L02 and the human cervical cancer cell lines HeLa. The EOs from the upper region of E. canadensis were prepared, its components were identified by GC/MS. Cell viability, cell morphology observation, AO/EB dual fluorescence staining assay, flow cytometry, mitochondrial membrane potential, western blot, caspase inhibitor test, and oxidative stress tests were used to investigate the impact of the EOs on HeLa cells. Network pharmacological analysis was employed to study the potential mechanism of the EOs in the treatment of cervical cancer. According to the findings, the EOs had 21 chemical components, of which limonene made up 65.68 %. After being exposed to the EOs, the cell viability of HeLa and L02 dramatically declined. The inhibition of EOs was more effective than that of limonene when used in an amount equivalent to that in the EOs. L02 cells were less susceptible to the cytotoxicity of EOs than HeLa cells were. Furthermore, EOs altered the cell cycle in HeLa cells and caused oxidative stress and apoptosis. Compared with the control group, the reactive oxygen species (ROS) levels increased in HeLa cells at first and then decreased, total superoxide dismutase (SOD) and catalase (CAT) activities in HeLa cells significantly decreased. G1 phase cells decreased whereas G2/M phase cells increased. The rate of apoptosis rose. Reduced mitochondrial membrane potential and Caspase-3, À 9, and À 12 protein expression were both observed. Nerolidol, dextroparaffinone, and α-pinene were shown to be the primary components for the suppression of HeLa cells, according to the results of the prediction of pharmacologic targets. In conclusion, findings of this study indicated the EOs may have the potential to curb the growth of cervical cancer cells. Further research is needed to explore the in vivo effect of EOs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.