Search

AbstractMonitoring online customer reviews is important for business organizations to measure customer satisfaction and better manage their reputations. In this paper, we propose a novel dynamic Brand-Topic Model (dBTM) which is able to automatically detect and track brand-associated sentiment scores and polarity-bearing topics from product reviews organized in temporally ordered time intervals. dBTM models the evolution of the latent brand polarity scores and the topic-word distributions over time by Gaussian state space models. It also incorporates a meta learning strategy to control the update of the topic-word distribution in each time interval in order to ensure smooth topic transitions and better brand score predictions. It has been evaluated on a dataset constructed from MakeupAlley reviews and a hotel review dataset. Experimental results show that dBTM outperforms a number of competitive baselines in brand ranking, achieving a good balance of topic coherence and uniqueness, and extracting well-separated polarity-bearing topics across time intervals.1

IntroductionSince the second half of the 20th century, Aedes albopictus, a vector for more than 20 arboviruses, has spread worldwide. Aedes albopictus is the main vector of infectious diseases transmitted by Aedes mosquitoes in China, and it has caused concerns regarding public health. A comprehensive understanding of the spatial genetic structure of this vector species at a genomic level is essential for effective vector control and the prevention of vector-borne diseases.MethodsDuring 2016–2018, adult female Ae. albopictus mosquitoes were collected from eight different geographical locations across China. Restriction site-associated DNA sequencing (RAD-seq) was used for high-throughput identification of single nucleotide polymorphisms (SNPs) and genotyping of the Ae. albopictus population. The spatial genetic structure was analyzed and compared to those exhibited by mitochondrial cytochrome c oxidase subunit 1 (cox1) and microsatellites in the Ae. albopictus population.ResultsA total of 9,103 genome-wide SNP loci in 101 specimens and 32 haplotypes of cox1 in 231 specimens were identified in the samples from eight locations in China. Principal component analysis revealed that samples from Lingshui and Zhanjiang were more genetically different than those from the other locations. The SNPs provided a better resolution and stronger signals for novel spatial population genetic structures than those from the cox1 data and a set of previously genotyped microsatellites. The fixation indexes from the SNP dataset showed shallow but significant genetic differentiation in the population. The Mantel test indicated a positive correlation between genetic distance and geographical distance. However, the asymmetric gene flow was detected among the populations, and it was higher from south to north and west to east than in the opposite directions.ConclusionsThe genome-wide SNPs revealed seven gene pools and fine spatial genetic structure of the Ae. albopictus population in China. The RAD-seq approach has great potential to increase our understanding of the spatial dynamics of population spread and establishment, which will help us to design new strategies for controlling vectors and mosquito-borne diseases.

Recent years have witnessed increasing interest in developing interpretable models in Natural Language Processing (NLP). Most existing models aim at identifying input features such as words or phrases important for model predictions. Neural models developed in NLP, however, often compose word semantics in a hierarchical manner. As such, interpretation by words or phrases only cannot faithfully explain model decisions in text classification. This article proposes a novel Hierarchical Interpretable Neural Text classifier, called HINT, which can automatically generate explanations of model predictions in the form of label-associated topics in a hierarchical manner. Model interpretation is no longer at the word level, but built on topics as the basic semantic unit. Experimental results on both review datasets and news datasets show that our proposed approach achieves text classification results on par with existing state-of-the-art text classifiers, and generates interpretations more faithful to model predictions and better understood by humans than other interpretable neural text classifiers.1

The susceptibility of Asian tiger mosquitoes to DENV-2 in different seasons was observed in simulated field environments as a reference to design dengue fever control strategies in Guangzhou. The life table experiments of mosquitoes in four seasons were carried out in the field. The susceptibility of Ae. albopictus to dengue virus was observed in both environments in Guangzhou in summer and winter. Ae. albopictus was infected with dengue virus by oral feeding. On day 7 and 14 after infection, the viral load in the head, ovary, and midgut of the mosquito was detected using real-time fluorescent quantitative PCR. Immune-associated gene expression in infected mosquitoes was performed using quantitative real-time reverse transcriptase PCR. The hatching rate and pupation rate of Ae. albopictus larvae in different seasons differed significantly. The winter hatching rate of larvae was lower than that in summer, and the incubation time was longer than in summer. In the winter field environment, Ae. albopictus still underwent basic growth and development processes. Mosquitoes in the simulated field environment were more susceptible to DENV-2 than those in the simulated laboratory environment. In the midgut, viral RNA levels on day 7 in summer were higher than those on day 7 in winter (F = 14.459, P = 0.01); ovarian viral RNA levels on day 7 in summer were higher than those on day 7 in winter (F = 8.656, P < 0.001), but there was no significant difference in the viral load at other time points (P > 0.05). Dicer-2 mRNA expression on day 7 in winter was 4.071 times than that on day 7 in summer: the viral load and Dicer-2 expression correlated moderately. Ae. albopictus could still develop and transmit dengue virus in winter in Guangzhou. Mosquitoes under simulated field conditions were more susceptible to DENV-2 than those under simulated laboratory conditions.

Abstract Background Aedes albopictus is a highly invasive mosquito species and a major vector of numerous viral pathogens. Many recent dengue fever outbreaks in China have been caused solely by the vector. Mapping of the potential distribution ranges of Ae. albopictus is crucial for epidemic preparedness and the monitoring of vector populations for disease control. Climate is a key factor influencing the distribution of the species. Despite field studies indicating seasonal population variations, very little modeling work has been done to analyze how environmental conditions influence the seasonality of Ae. albopictus. The aim of the present study was to develop a model based on available observations, climatic and environmental data, and machine learning methods for the prediction of the potential seasonal ranges of Ae. albopictus in China. Methods We collected comprehensive up-to-date surveillance data in China, particularly records from the northern distribution margin of Ae. albopictus. All records were assigned long-term (1970–2000) climatic data averages based on the WorldClim 2.0 data set. Machine learning regression tree models were developed using a 10-fold cross-validation method to predict the potential seasonal (or monthly) distribution ranges of Ae. albopictus in China at high resolution based on environmental conditions. The models were assessed based on sensitivity, specificity, and accuracy, using area under curve (AUC). WorldClim 2.0 and climatic and environmental data were used to produce environmental conduciveness (probability) prediction surfaces. Predicted probabilities were generated based on the averages of the 10 models. Results During 1998–2017, Ae. albopictus was observed at 200 out of the 242 localities surveyed. In addition, at least 15 new Ae. albopictus occurrence sites lay outside the potential ranges that have been predicted using models previously. The average accuracy was 98.4% (97.1–99.5%), and the average AUC was 99.1% (95.6–99.9%). The predicted Ae. albopictus distribution in winter (December–February) was limited to a small subtropical-tropical area of China, and Ae. albopictus was predicted to occur in northern China only during the short summer season (usually June–September). The predicted distribution areas in summer could reach northeastern China bordering Russia and the eastern part of the Qinghai-Tibet Plateau in southwestern China. Ae. albopictus could remain active in expansive areas from central to southern China in October and November. Conclusions Climate and environmental conditions are key factors influencing the seasonal distribution of Ae. albopictus in China. The areas predicted to potentially host Ae. albopictus seasonally in the present study could reach northeastern China and the eastern slope of the Qinghai-Tibet Plateau. Our results present new evidence and suggest the expansion of systematic vector population monitoring activities and regular re-assessment of epidemic risk potential.

Abstract Background The Asian tiger mosquito, Aedes albopictus, is one of the 100 worst invasive species in the world and the vector for several arboviruses including dengue, Zika and chikungunya viruses. Understanding the population spatial genetic structure, migration, and gene flow of vector species is critical to effectively preventing and controlling vector-borne diseases. Little is known about the population structure and genetic differentiation of native Ae. albopictus in China. The aim of this study was to examine the patterns of the spatial genetic structures of native Ae. albopictus populations, and their relationship to dengue incidence, on a large geographical scale. Methods During 2016–2018, adult female Ae. albopictus mosquitoes were collected by human landing catch (HLC) or human-bait sweep-net collections in 34 localities across China. Thirteen microsatellite markers were used to examine the patterns of genetic diversity, population structure, and gene flow among native Ae. albopictus populations. The correlation between population genetic indices and dengue incidence was also examined. Results A total of 153 distinct alleles were identified at the 13 microsatellite loci in the tested populations. All loci were polymorphic, with the number of distinct alleles ranging from eight to sixteen. Genetic parameters such as PIC, heterozygosity, allelic richness and fixation index (F ST) revealed highly polymorphic markers, high genetic diversity, and low population genetic differentiation. In addition, Bayesian analysis of population structure showed two distinct genetic groups in southern-western and eastern-central-northern China. The Mantel test indicated a positive correlation between genetic distance and geographical distance (R 2 = 0.245, P = 0.01). STRUCTURE analysis, PCoA and GLS interpolation analysis indicated that Ae. albopictus populations in China were regionally clustered. Gene flow and relatedness estimates were generally high between populations. We observed no correlation between population genetic indices of microsatellite loci in Ae. albopictus populations and dengue incidence. Conclusion Strong gene flow probably assisted by human activities inhibited population differentiation and promoted genetic diversity among populations of Ae. albopictus. This may represent a potential risk of rapid spread of mosquito-borne diseases. The spatial genetic structure, coupled with the association between genetic indices and dengue incidence, may have important implications for understanding the epidemiology, prevention, and control of vector-borne diseases.

Neural attention mechanism has achieved many successes in various tasks in natural language processing. However, existing neural attention models based on a densely connected network are loosely related to the attention mechanism found in psychology and neuroscience. Motivated by the finding in neuroscience that human possesses the template-searching attention mechanism, we propose to use convolution operation to simulate attentions and give a mathematical explanation of our neural attention model. We then introduce a new network architecture, which combines a recurrent neural network with our convolution-based attention model and further stacks an attention-based neural model to build a hierarchical sentiment classification model. The experimental results show that our proposed models can capture salient parts of the text to improve the performance of sentiment classification at both the sentence level and the document level.

BACKGROUND:Aedes albopictus is a major vector for several tropical infectious diseases. Characterization of Ae. albopictus development under natural conditions is crucial for monitoring vector population expansion, dengue virus transmission, and disease outbreak preparedness. METHODS:This study employed mosquito traits as a proxy to understanding life-table traits in mosquitoes using a semi-field study. Ae. albopictus larval and adult life-table experiments were conducted using microcosms under semi-field conditions in Guangzhou. Stage-specific development times and survivorship rates were determined and compared under semi-field conditions in different seasons from early summer (June) to winter (January), to determine the lower temperature limit for larval development and adult survivorship and reproductivity. RESULTS:The average egg- hatching rate was 60.1%, with the highest recorded in October (77.1%; mid-autumn). The larval development time was on average 13.2 days (range, 8.5-24.1 days), with the shortest time observed in September(8.7 days; early autumn) and longest in November (22.8 days). The pupation rates of Ae. albopictus larvae were on average 88.9% (range, 81.6-93.4%); they were stable from June to September but decreased from October to November. The adult emergence rates were on average 82.5% (range, 76.8-87.9%) and decreased from July to November. The median survival time of Ae. albopictus adults was on average 7.4 (range, 4.5-9.8), with the shortest time recorded in September. The average lifetime egg mass under semi-field conditions was 37.84 eggs/female. The larvae could develop into adults at temperatures as low as 12.3°C, and the adults could survive for 30.0 days at 16.3°C and still produce eggs. Overall, correlation analysis found that mean temperature and relative humidity were variables significantly affecting larval development and adult survivorship. CONCLUSION:Ae. albopictus larvae could develop and emerge and the adults could survive and produce eggs in early winter in Guangzhou. The major impact of changes in ambient temperature, relative humidity, and light intensity was on the egg hatching rates, adult survival time, and egg mass production, rather than on pupation or adult emergence rates.

Abstract Background Prediction of DNA-binding residue is important for understanding the protein-DNA recognition mechanism. Many computational methods have been proposed for the prediction, but most of them do not consider the relationships of evolutionary information between residues. Results In this paper, we first propose a novel residue encoding method, referred to as the Position Specific Score Matrix (PSSM) Relation Transformation (PSSM-RT), to encode residues by utilizing the relationships of evolutionary information between residues. PDNA-62 and PDNA-224 are used to evaluate PSSM-RT and two existing PSSM encoding methods by five-fold cross-validation. Performance evaluations indicate that PSSM-RT is more effective than previous methods. This validates the point that the relationship of evolutionary information between residues is indeed useful in DNA-binding residue prediction. An ensemble learning classifier (EL_PSSM-RT) is also proposed by combining ensemble learning model and PSSM-RT to better handle the imbalance between binding and non-binding residues in datasets. EL_PSSM-RT is evaluated by five-fold cross-validation using PDNA-62 and PDNA-224 as well as two independent datasets TS-72 and TS-61. Performance comparisons with existing predictors on the four datasets demonstrate that EL_PSSM-RT is the best-performing method among all the predicting methods with improvement between 0.02–0.07 for MCC, 4.18–21.47% for ST and 0.013–0.131 for AUC. Furthermore, we analyze the importance of the pair-relationships extracted by PSSM-RT and the results validates the usefulness of PSSM-RT for encoding DNA-binding residues. Conclusions We propose a novel prediction method for the prediction of DNA-binding residue with the inclusion of relationship of evolutionary information and ensemble learning. Performance evaluation shows that the relationship of evolutionary information between residues is indeed useful in DNA-binding residue prediction and ensemble learning can be used to address the data imbalance issue between binding and non-binding residues. A web service of EL_PSSM-RT ( http://hlt.hitsz.edu.cn:8080/PSSM-RT_SVM/ ) is provided for free access to the biological research community.

In recent years, there has been an increasing interest in learning a distributed representation of word sense. Traditional context clustering based models usually require careful tuning of model parameters, and typically perform worse on infrequent word senses. This paper presents a novel approach which addresses these limitations by first initializing the word sense embeddings through learning sentence-level embeddings from WordNet glosses using a convolutional neural networks. The initialized word sense embeddings are used by a context clustering based model to generate the distributed representations of word senses. Our learned representations outperform the publicly available embeddings on half of the metrics in the word similarity task, 6 out of 13 sub tasks in the analogical reasoning task, and gives the best overall accuracy in the word sense effect classification task, which shows the effectiveness of our proposed distributed distribution learning model.

Natural language understanding is to specify a computational model that maps sentences to their semantic mean representation. In this paper, we propose a novel framework to train the statistical models without using expensive fully annotated data. In particular, the input of our framework is a set of sentences labeled with abstract semantic annotations. These annotations encode the underlying embedded semantic structural relations without explicit word/semantic tag alignment. The proposed framework can automatically induce derivation rules that map sentences to their semantic meaning representations. The learning framework is applied on two statistical models, the conditional random fields (CRFs) and the hidden Markov support vector machines (HM-SVMs). Our experimental results on the DARPA communicator data show that both CRFs and HM-SVMs outperform the baseline approach, previously proposed hidden vector state (HVS) model which is also trained on abstract semantic annotations. In addition, the proposed framework shows superior performance than two other baseline approaches, a hybrid framework combining HVS and HM-SVMs and discriminative training of HVS, with a relative error reduction rate of about 25% and 15% being achieved in F-measure.