Deep joint learning for language recognition.

Deep learning methods for language recognition have achieved promising performance. However, most of the studies focus on frameworks for single types of acoustic features and single tasks. In this paper, we propose the deep joint learning strategies based on the Multi-Feature (MF) and Multi-Task (MT) models. First, we investigate the efficiency of integrating multiple acoustic features and explore two kinds of training constraints, one is introducing auxiliary classification constraints with adaptive weights for loss functions in feature encoder sub-networks, and the other option is introducing the Canonical Correlation Analysis (CCA) constraint to maximize the correlation of different feature representations. Correlated speech tasks, such as phoneme recognition, are applied as auxiliary tasks in order to learn related information to enhance the performance of language recognition. We analyze phoneme-aware information from different learning strategies, like joint learning on the frame-level, adversarial learning on the segment-level, and the combination mode. In addition, we present the Language-Phoneme embedding extraction structure to learn and extract language and phoneme embedding representations simultaneously. We demonstrate the effectiveness of the proposed approaches with experiments on the Oriental Language Recognition (OLR) data sets. Experimental results indicate that joint learning on the multi-feature and multi-task models extracts instinct feature representations for language identities and improves the performance, especially in complex challenges, such as cross-channel or open-set conditions.

Systematic analysis and prediction of type IV secreted effector proteins by machine learning approaches.

In the course of infecting their hosts, pathogenic bacteria secrete numerous effectors, namely, bacterial proteins that pervert host cell biology. Many Gram-negative bacteria, including context-dependent human pathogens, use a type IV secretion system (T4SS) to translocate effectors directly into the cytosol of host cells. Various type IV secreted effectors (T4SEs) have been experimentally validated to play crucial roles in virulence by manipulating host cell gene expression and other processes. Consequently, the identification of novel effector proteins is an important step in increasing our understanding of host-pathogen interactions and bacterial pathogenesis. Here, we train and compare six machine learning models, namely, Naïve Bayes (NB), K-nearest neighbor (KNN), logistic regression (LR), random forest (RF), support vector machines (SVMs) and multilayer perceptron (MLP), for the identification of T4SEs using 10 types of selected features and 5-fold cross-validation. Our study shows that: (1) including different but complementary features generally enhance the predictive performance of T4SEs; (2) ensemble models, obtained by integrating individual single-feature models, exhibit a significantly improved predictive performance and (3) the 'majority voting strategy' led to a more stable and accurate classification performance when applied to predicting an ensemble learning model with distinct single features. We further developed a new method to effectively predict T4SEs, Bastion4 (Bacterial secretion effector predictor for T4SS), and we show our ensemble classifier clearly outperforms two recent prediction tools. In summary, we developed a state-of-the-art T4SE predictor by conducting a comprehensive performance evaluation of different machine learning algorithms along with a detailed analysis of single- and multi-feature selections.

Precise prediction model and simplified scoring system for sustained combined response to interferon-alpha.

AIM: To establish a predictive algorithm which may serve for selecting optimal candidates for interferon-alpha (IFN-alpha) treatment. METHODS: A total of 474 IFN-alpha treated hepatitis B virus e antigen (HBeAg)-positive patients were enrolled in the present study. The patients' baseline characteristics, such as age, gender, blood tests, activity grading (G) of intrahepatic inflammation, score (S) of liver fibrosis, hepatitis B virus (HBV) DNA and genotype were evaluated; therapy duration and response of each patient at the 24th wk after cessation of IFN-alpha treatment were also recorded. A predictive algorithm and scoring system for a sustained combined response (CR) to IFN-alpha therapy were established. About 10% of the patients were randomly drawn as the test set. Responses to IFN-alpha therapy were divided into CR, partial response (PR) and non-response (NR). The mixed set of PR and NR was recorded as PR+NR. RESULTS: Stratified by therapy duration, the most significant baseline predictive factors were alanine aminotransferase (ALT), HBV DNA level, aspartate aminotransferase (AST), HBV genotype, S, G, age and gender. According to the established model, the accuracies for sustained CR and PR+NR, respectively, were 86.4% and 93.0% for the training set, 81.5% and 91.0% for the test set. For the scoring system, the sensitivity and specificity were 78.8% and 80.6%, respectively. There were positive correlations between ALT and AST, and G and S, respectively. CONCLUSION: With these models, practitioners may be able to propose individualized decisions that have an integrated foundation on both evidence-based medicine and personal characteristics.