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.

This research was conducted within the EU projects e-SENSE and SENSEI. Copyright c 2010 by Yang Zhang, Enschede, The Netherlands. All rights reserved. No part of this book may be reproduced or transmitted, in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without the prior written permission of the author.Printed by Wöhrmann Print Service. AbstractThe generation of wireless sensor networks (WSNs) makes human beings observe and reason about the physical environment better, easier, and faster. The wireless sensor nodes equipped with sensing, processing, wireless communication and actuation capabilities can be densely deployed in a wide geographical area and measure various parameters continuously from the physical world. Compared with traditional environmental sensing technologies, such densely deployed WSNs enable collection of fine-grained high spatial and temporal resolution data with less installation, maintenance, and operation costs.However, raw sensor observations often have low data quality and reliability due to both internal and external factors including low quality of cheap sensors, dynamicity of network conditions, and harshness of the deployment environment. Use of low quality sensor data in any data analysis and decision making process will not only negatively impact analysis results and decisions made but also waste huge amount of valuable and limited network resources such as energy, as many incorrect values are transmitted. Low quality sensor data also prevents WSNs to fulfill their promises in terms of reliable real-time situation-awareness, as the low quality sensor data may generate large number of false alarms.Motivated by the need to improve quality of data analysis and decision making, enhance efficiency of using WSNs resources by preventing unnecessary transmission of erroneous sensor observations, and increase effectiveness of monitoring and situation-awareness capabilities of the WSNs, in this thesis we focus on online identification of outliers whenever and wherever they occur. Outliers in WSNs are those observations that represent erroneous values (errors) or indicate particular phenomenal changes (events). Our outlier detection techniques, which are based on distributed in-network data processing, identify sensor observations that do not conform to normal behavior of sensor data without using a pre-defined threshold or triggering conditions.Our main research objective is to design and implement effective and efficient outlier detection techniques for WSNs to identify outliers in an online and disv tributed manner and distinguish between errors and events with high accuracy and low false alarm, while maintaining the communication, computation and memory complexity low. Main contributions of this thesis can be summarized as: 3. Statistical-Based outlier detection techniques for WSNs. We take two approaches in designing our outlier detection techniques. One approach originates fr...

The main challenge faced by outlier detection techniques designed for wireless sensor networks is achieving high detection rate and low false alarm rate while maintaining the resource consumption in the network to a minimum. In this paper, we propose an online outlier detection technique with low computational complexity and memory usage based on an unsupervised centered quarter-sphere support vector machine for real-time environmental monitoring applications of wireless sensor networks. The proposed approach is completely local and thus saves communication overhead and scales well with increase of nodes deployed. We take advantage of spatial correlations that exist in sensor data of adjacent nodes to reduce the false alarm rate in real-time. Experiments with both synthetic and real data collected from the Intel Berkeley Research Laboratory show that our technique achieves better mining performance in terms of parameter selection using different kernel functions compared to an earlier offline outlier detection technique designed for wireless sensor networks.

Hashtags in online social networks have gained tremendous popularity during the past five years. The resulting large quantity of data has provided a new lens into modern society. Previously, researchers mainly rely on data collected from Twitter to study either a certain type of hashtags or a certain property of hashtags. In this paper, we perform the first large-scale empirical analysis of hashtags shared on Instagram, the major platform for hashtag-sharing. We study hashtags from three different dimensions including the temporal-spatial dimension, the semantic dimension, and the social dimension. Extensive experiments performed on three large-scale datasets with more than 7 million hashtags in total provide a series of interesting observations. First, we show that the temporal patterns of hashtags can be categorized into four different clusters, and people tend to share fewer hashtags at certain places and more hashtags at others. Second, we observe that a non-negligible proportion of hashtags exhibit large semantic displacement. We demonstrate hashtags that are more uniformly shared among users, as quantified by the proposed hashtag entropy, are less prone to semantic displacement. In the end, we propose a bipartite graph embedding model to summarize users' hashtag profiles, and rely on these profiles to perform friendship prediction. Evaluation results show that our approach achieves an effective prediction with AUC (area under the ROC curve) above 0.8 which demonstrates the strong social signals possessed in hashtags.

Conventional supervised binary classification algorithms have been widely applied to address significant research questions using biological and biomedical data. This classification scheme requires two fully labeled classes of data (e.g. positive and negative samples) to train a classification model. However, in many bioinformatics applications, labeling data is laborious, and the negative samples might be potentially mislabeled due to the limited sensitivity of the experimental equipment. The positive unlabeled (PU) learning scheme was therefore proposed to enable the classifier to learn directly from limited positive samples and a large number of unlabeled samples (i.e. a mixture of positive or negative samples). To date, several PU learning algorithms have been developed to address various biological questions, such as sequence identification, functional site characterization and interaction prediction. In this paper, we revisit a collection of 29 state-of-the-art PU learning bioinformatic applications to address various biological questions. Various important aspects are extensively discussed, including PU learning methodology, biological application, classifier design and evaluation strategy. We also comment on the existing issues of PU learning and offer our perspectives for the future development of PU learning applications. We anticipate that our work serves as an instrumental guideline for a better understanding of the PU learning framework in bioinformatics and further developing next-generation PU learning frameworks for critical biological applications.