Search

Additional file 6. Table S10. Top 5 cities of cystic and alveolar echinococcosis prevalence in China. Table S11. Top 10 counties of cystic and alveolar echinococcosis prevalence in China.

Additional file 5. Table S1. Demographic characteristics of human Echinococcosis in Inner Mongolia. Table S2. Demographic characteristics of human Echinococcosis in Sichuan. Table S3. Demographic characteristics of human Echinococcosis in Yunnan. Table S4. Demographic characteristics of human Echinococcosis in Tibet. Table S5. Demographic characteristics of human Echinococcosis in Shaanxi. Table S6. Demographic characteristics of human Echinococcosis in Gansu. Table S7. Demographic characteristics of human Echinococcosis in Qinghai. Table S8. Demographic characteristics of human Echinococcosis in Ningxia. Table S9. Demographic characteristics of human Echinococcosis in Xinjiang.

Additional file 1: Table S1. Genus/Species compositions of Anopheles mosquitoes trapped by the different methods in Siem Pang, Loum and Pangkhom villages.

Additional file 7. Table S12. Statistic results of GLM model for cystic and alveolar echinococcosis.

Additional file 1. Questionnaire 1. Household basic information questionnaire. Questionnaire 2. Knowledge for echinococcosis prevention and control.

Additional file 2: Table S2. The percentage of mosquitoes with Plasmodium-positive by the different methods in Asian countries.

Additional file 2. Figure S1. The spatial distribution of environmental, biological and social factors adopted in this study. Figure S2. Correlation and variance inflation factorof variables adopted in cystic echinococcosis modeling. Figure S3. Correlation and variance inflation factorof variables adopted in alveolar echinococcosis modeling. Figure S4. Sex-age specific distribution of cystic echinococcosisand alveolar echinococcosisat nation level. Figure S5. Sex-age specific distribution of cystic echinococcosisand alveolar echinococcosisby province.

Pseudo Labeling is a technique used to improve the performance of semi-supervised Graph Neural Networks (GNNs) by generating additional pseudo-labels based on confident predictions. However, the quality of generated pseudo-labels has been a longstanding concern due to the sensitivity of the classification objective with respect to the given labels. To avoid the untrustworthy classification supervision indicating ``a node belongs to a specific class,'' we favor the fault-tolerant contrasting supervision demonstrating ``two nodes do not belong to the same class.'' Thus, the problem of generating high-quality pseudo-labels is then transformed into a relaxed version, i.e., identifying reliable negative pairs. To achieve this, we propose a general framework for GNNs, termed Pseudo Contrastive Learning (PCL). It separates two nodes whose positive and negative pseudo-labels target the same class. To incorporate topological knowledge into learning, we devise a topologically weighted contrastive loss that spends more effort separating negative pairs with smaller topological distances. Experimentally, we apply PCL to various GNNs, which consistently outperform their counterparts using other popular general techniques on five real-world graphs., Comment: Under Review

To create a large amount of training labels for machine learning models effectively and efficiently, researchers have turned to Weak Supervision (WS), which uses programmatic labeling sources rather than manual annotation. Existing works of WS for binary classification typically assume the presence of labeling sources that are able to assign both positive and negative labels to data in roughly balanced proportions. However, for many tasks of interest where there is a minority positive class, negative examples could be too diverse for developers to generate indicative labeling sources. Thus, in this work, we study the application of WS on binary classification tasks with positive labeling sources only. We propose WEAPO, a simple yet competitive WS method for producing training labels without negative labeling sources. On 10 benchmark datasets, we show WEAPO achieves the highest averaged performance in terms of both the quality of synthesized labels and the performance of the final classifier supervised with these labels. We incorporated the implementation of \method into WRENCH, an existing benchmarking platform., CIKM 2022 (short)

Additional file 6: Table S4. Mean intra- and interspecific K2P distances of the ITS2 sequence (a) and COII sequence (b) in 12 Anopheles species. The numbers of intraspecific distances are shown in boldface for clarity. Numbers underlined indicate the highest intraspecific distance and the lowest interspecific distance. Abbreviations: n, no. of sequences; na, not applicable; arg., An. argyropus; nig., An. nigerrimus; nit., An. nitidus; ped., An. peditaeniatus; sin., An. sinensis; niv., An. nivipes; tes., An. tessellatus; dir., An. dirus; mac., An. maculatus; phi., An. philippinensis; koc., An. kochi; vag., An. vagus.

Additional file 9: Figure S5. Phylogenetic tree based on 18 ITS2 sequences (a) and 15 COII sequences (b) of Anopheles tessellatus from GenBank and our original data. Bootstrap values (1000 replicates) of maximum likelihood analyses are shown above/below the main lineages. Lineage designation is indicated on the right. Bars represent 2.0 substitutions per site based on ITS2 and 0.6 substitutions per site based on COII.

Structured pruning techniques have achieved great compression performance on convolutional neural networks for image classification task. However, the majority of existing methods are weight-oriented, and their pruning results may be unsatisfactory when the original model is trained poorly. That is, a fully-trained model is required to provide useful weight information. This may be time-consuming, and the pruning results are sensitive to the updating process of model parameters. In this paper, we propose a metric named Average Filter Information Entropy (AFIE) to measure the importance of each filter. It is calculated by three major steps, i.e., low-rank decomposition of the "input-output" matrix of each convolutional layer, normalization of the obtained eigenvalues, and calculation of filter importance based on information entropy. By leveraging the proposed AFIE, the proposed framework is able to yield a stable importance evaluation of each filter no matter whether the original model is trained fully. We implement our AFIE based on AlexNet, VGG-16, and ResNet-50, and test them on MNIST, CIFAR-10, and ImageNet, respectively. The experimental results are encouraging. We surprisingly observe that for our methods, even when the original model is only trained with one epoch, the importance evaluation of each filter keeps identical to the results when the model is fully-trained. This indicates that the proposed pruning strategy can perform effectively at the beginning stage of the training process for the original model.

Recent self-supervised pre-training methods on Heterogeneous Information Networks (HINs) have shown promising competitiveness over traditional semi-supervised Heterogeneous Graph Neural Networks (HGNNs). Unfortunately, their performance heavily depends on careful customization of various strategies for generating high-quality positive examples and negative examples, which notably limits their flexibility and generalization ability. In this work, we present SHGP, a novel Self-supervised Heterogeneous Graph Pre-training approach, which does not need to generate any positive examples or negative examples. It consists of two modules that share the same attention-aggregation scheme. In each iteration, the Att-LPA module produces pseudo-labels through structural clustering, which serve as the self-supervision signals to guide the Att-HGNN module to learn object embeddings and attention coefficients. The two modules can effectively utilize and enhance each other, promoting the model to learn discriminative embeddings. Extensive experiments on four real-world datasets demonstrate the superior effectiveness of SHGP against state-of-the-art unsupervised baselines and even semi-supervised baselines. We release our source code at: https://github.com/kepsail/SHGP., Comment: Accepted by NeurIPS 2022

Additional file 5: Figure S2. Phylogenetic tree based on 1264 COII sequences (507 haplotypes) from GenBank and our original data. Bootstrap values (1000 replicates) of maximum likelihood analyses are shown above/below the main lineages. Lineage designation is indicated on the right. Bars represent 5.0 substitutions per site based on COII.

Additional file 8: Figure S4. ITS2 and COII sequence divergence. Each dot represents a species, with interspecific distance on the y-axis and intraspecific distance on the x-axis. The minimum interspecific (intergroup) divergence is plotted against the maximum intraspecific divergence. Red and green dots indicate the ITS2 and COII sequence divergence of 12 species including An. argyropus, An. nigerrimus, An. nitidus, An. peditaeniatus, An. sinensis, An. nivipes, An. tessellatus, An. dirus, An. maculatus, An. philippinensis, An. kochi and An. vagus.

Responsing with image has been recognized as an important capability for an intelligent conversational agent. Yet existing works only focus on exploring the multimodal dialogue models which depend on retrieval-based methods, but neglecting generation methods. To fill in the gaps, we first present a multimodal dialogue generation model, which takes the dialogue history as input, then generates a textual sequence or an image as response. Learning such a model often requires multimodal dialogues containing both texts and images which are difficult to obtain. Motivated by the challenge in practice, we consider multimodal dialogue generation under a natural assumption that only limited training examples are available. In such a low-resource setting, we devise a novel conversational agent, Divter, in order to isolate parameters that depend on multimodal dialogues from the entire generation model. By this means, the major part of the model can be learned from a large number of text-only dialogues and text-image pairs respectively, then the whole parameters can be well fitted using the limited training examples. Extensive experiments demonstrate our method achieves state-of-the-art results in both automatic and human evaluation, and can generate informative text and high-resolution image responses., Comment: Accepted to ACL 2022 Main Conference

Session-based recommendation aims to predict a user's next action based on previous actions in the current session. The major challenge is to capture authentic and complete user preferences in the entire session. Recent work utilizes graph structure to represent the entire session and adopts Graph Neural Network to encode session information. This modeling choice has been proved to be effective and achieved remarkable results. However, most of the existing studies only consider each item within the session independently and do not capture session semantics from a high-level perspective. Such limitation often leads to severe information loss and increases the difficulty of capturing long-range dependencies within a session. Intuitively, compared with individual items, a session snippet, i.e., a group of locally consecutive items, is able to provide supplemental user intents which are hardly captured by existing methods. In this work, we propose to learn multi-granularity consecutive user intent unit to improve the recommendation performance. Specifically, we creatively propose Multi-granularity Intent Heterogeneous Session Graph which captures the interactions between different granularity intent units and relieves the burden of long-dependency. Moreover, we propose the Intent Fusion Ranking module to compose the recommendation results from various granularity user intents. Compared with current methods that only leverage intents from individual items, IFR benefits from different granularity user intents to generate more accurate and comprehensive session representation, thus eventually boosting recommendation performance. We conduct extensive experiments on five session-based recommendation datasets and the results demonstrate the effectiveness of our method.

One of the key challenges in Neural Architecture Search (NAS) is to efficiently rank the performances of architectures. The mainstream assessment of performance rankers uses ranking correlations (e.g., Kendall's tau), which pay equal attention to the whole space. However, the optimization goal of NAS is identifying top architectures while paying less attention on other architectures in the search space. In this paper, we show both empirically and theoretically that Normalized Discounted Cumulative Gain (NDCG) is a better metric for rankers. Subsequently, we propose a new algorithm, AceNAS, which directly optimizes NDCG with LambdaRank. It also leverages weak labels produced by weight-sharing NAS to pre-train the ranker, so as to further reduce search cost. Extensive experiments on 12 NAS benchmarks and a large-scale search space demonstrate that our approach consistently outperforms SOTA NAS methods, with up to 3.67% accuracy improvement and 8x reduction on search cost., Comment: Code: https://github.com/ultmaster/AceNAS

Creating labeled training sets has become one of the major roadblocks in machine learning. To address this, recent \emph{Weak Supervision (WS)} frameworks synthesize training labels from multiple potentially noisy supervision sources. However, existing frameworks are restricted to supervision sources that share the same output space as the target task. To extend the scope of usable sources, we formulate Weak Indirect Supervision (WIS), a new research problem for automatically synthesizing training labels based on indirect supervision sources that have different output label spaces. To overcome the challenge of mismatched output spaces, we develop a probabilistic modeling approach, PLRM, which uses user-provided label relations to model and leverage indirect supervision sources. Moreover, we provide a theoretically-principled test of the distinguishability of PLRM for unseen labels, along with a generalization bound. On both image and text classification tasks as well as an industrial advertising application, we demonstrate the advantages of PLRM by outperforming baselines by a margin of 2%-9%., Comment: ICLR 2022

This contribution is concerned with price optimisation of the new business for a non-life product. Due to high competition in the insurance market, non-life insurers are interested in increasing their conversion rates on new business based on some profit level. In this respect, we consider the competition in the market to model the probability of accepting an offer for a specific customer. We study two optimisation problems relevant for the insurer and present some algorithmic solutions for both continuous and discrete case. Finally, we provide some applications to a motor insurance dataset.

Cruise travel has become increasingly popular, and the number of cruisers is growing rapidly. China plays a dominant role in worldwide tourism owing to its large population and fast economic development. It is a potential major market for the cruise industry. The number of inbound and outbound Chinese travelers continues to increase, generating growing interest in cruise travel. Investigating what Chinese cruisers want from cruise travel has great significance; however, little research has been conducted concerning Chinese travelers' preferences in the cruise product domain. The purpose of the present study is to address this gap by investigating Chinese travelers' cruise product preferences. This study also features an exploratory analysis on motivational factors in a Chinese context. Theoretical and practical implications are discussed.

We investigated changes in the susceptibility of Plasmodium falciparum to artesunate in vitro using Rieckmann's microtest in Yunnan Province, China, during the period 1988 to 1999. Longitudinal surveillance studies in 1988, 1992 and 1999 revealed that the IC50s were 6.2, 7.2 and 20.7 nmol/L, respectively and mean concentrations completely inhibiting schizont formation (CIMC) were 37.8, 46.1 and 74.0 nmol/L, respectively. The IC50 and CIMC in 1999 were 3.3 and 2 times greater than in 1988. In cross-sectional tests from 1991 to 1993, the susceptibility of P. falciparum isolates from areas in the western and the southern parts of Yunnan Province were similar, but lower than in the south-eastern and central parts of the province. The results suggest that P. falciparum is generally susceptible to artemisinin derivatives but indicate a reduction in susceptibility during the study period in areas where the drugs have been used for a long time.