Automating aspects of biocuration through biomedical information extraction could significantly impact biomedical research by enabling greater biocuration throughput and improving the feasibility of a wider scope. An important step in biomedical information extraction systems is named entity recognition (NER), where mentions of entities such as proteins and diseases are located within natural-language text and their semantic type is determined. This step is critical for later tasks in an information extraction pipeline, including normalization and relationship extraction. BANNER is a benchmark biomedical NER system using linear-chain conditional random fields and the rich feature set approach. A case study with BANNER locating genes and proteins in biomedical literature is described. The first corpus for disease NER adequate for use as training data is introduced, and employed in a case study of disease NER. The first corpus locating adverse drug reactions (ADRs) in user posts to a health-related social website is also described, and a system to locate and identify ADRs in social media text is created and evaluated. The rich feature set approach to creating NER feature sets is argued to be subject to diminishing returns, implying that additional improvements may require more sophisticated methods for creating the feature set. This motivates the first application of multivariate feature selection with filters and false discovery rate analysis to biomedical NER, resulting in a feature set at least 3 orders of magnitude smaller than the set created by the rich feature set approach. Finally, two novel approaches to NER by modeling the semantics of token sequences are introduced. The first method focuses on the sequence content by using language models to determine whether a sequence resembles entries in a lexicon of entity names or text from an unlabeled corpus more closely. The second method models the distributional semantics of token sequences, determining the similarity between a potential mention and the token sequences from the training data by analyzing the contexts where each sequence appears in a large unlabeled corpus. The second method is shown to improve the performance of BANNER on multiple data sets.

The rapid urban expansion has greatly extended the physical boundary of our living area, along with a large number of POIs (points of interest) being developed. A POI is a specific location (e.g., hotel, restaurant, theater, mall) that a user may find useful or interesting. When exploring the city and neighborhood, the increasing number of POIs could enrich people's daily life, providing them with more choices of life experience than before, while at the same time also brings the problem of "curse of choices", resulting in the difficulty for a user to make a satisfied decision on "where to go" in an efficient way. Personalized POI recommendation is a task proposed on purpose of helping users filter out uninteresting POIs and reduce time in decision making, which could also benefit virtual marketing.

Developing POI recommender systems requires observation of human mobility w.r.t. real-world POIs, which is infeasible with traditional mobile data. However, the recent development of location-based social networks (LBSNs) provides such observation. Typical location-based social networking sites allow users to "check in" at POIs with smartphones, leave tips and share that experience with their online friends. The increasing number of LBSN users has generated large amounts of LBSN data, providing an unprecedented opportunity to study human mobility for personalized POI recommendation in spatial, temporal, social, and content aspects.

Different from recommender systems in other categories, e.g., movie recommendation in NetFlix, friend recommendation in dating websites, item recommendation in online shopping sites, personalized POI recommendation on LBSNs has its unique challenges due to the stochastic property of human mobility and the mobile behavior indications provided by LBSN information layout. The strong correlations between geographical POI information and other LBSN information result in three major human mobile properties, i.e., geo-social correlations, geo-temporal patterns, and geo-content indications, which are neither observed in other recommender systems, nor exploited in current POI recommendation. In this dissertation, we investigate these properties on LBSNs, and propose personalized POI recommendation models accordingly. The performance evaluated on real-world LBSN datasets validates the power of these properties in capturing user mobility, and demonstrates the ability of our models for personalized POI recommendation.

In many fields one needs to build predictive models for a set of related machine learning tasks, such as information retrieval, computer vision and biomedical informatics. Traditionally these tasks are treated independently and the inference is done separately for each task, which ignores important connections among the tasks. Multi-task learning aims at simultaneously building models for all tasks in order to improve the generalization performance, leveraging inherent relatedness of these tasks. In this thesis, I firstly propose a clustered multi-task learning (CMTL) formulation, which simultaneously learns task models and performs task clustering. I provide theoretical analysis to establish the equivalence between the CMTL formulation and the alternating structure optimization, which learns a shared low-dimensional hypothesis space for different tasks. Then I present two real-world biomedical informatics applications which can benefit from multi-task learning. In the first application, I study the disease progression problem and present multi-task learning formulations for disease progression. In the formulations, the prediction at each point is a regression task and multiple tasks at different time points are learned simultaneously, leveraging the temporal smoothness among the tasks. The proposed formulations have been tested extensively on predicting the progression of the Alzheimer's disease, and experimental results demonstrate the effectiveness of the proposed models. In the second application, I present a novel data-driven framework for densifying the electronic medical records (EMR) to overcome the sparsity problem in predictive modeling using EMR. The densification of each patient is a learning task, and the proposed algorithm simultaneously densify all patients. As such, the densification of one patient leverages useful information from other patients.

The widespread adoption of mobile devices gives rise to new opportunities and challenges for authentication mechanisms. Many traditional authentication mechanisms become unsuitable for smart devices. For example, while password is widely used on computers as user identity authentication, inputting password on small smartphone screen is error-prone and not convenient. In the meantime, there are emerging demands for new types of authentication. Proximity authentication is an example, which is not needed for computers but quite necessary for smart devices. These challenges motivate me to study and develop novel authentication mechanisms specific for smart devices.

In this dissertation, I am interested in the special authentication demands of smart devices and about to satisfy the demands. First, I study how the features of smart devices affect user identity authentications. For identity authentication domain, I aim to design a continuous, forge-resistant authentication mechanism that does not interrupt user-device interactions. I propose a mechanism that authenticates user identity based on the user's finger movement patterns. Next, I study a smart-device-specific authentication, proximity authentication, which authenticates whether two devices are in close proximity. For prox- imity authentication domain, I aim to design a user-friendly authentication mechanism that can defend against relay attacks. In addition, I restrict the authenticated distance to the scale of near field, i.e., a few centimeters. My first design utilizes a user's coherent two-finger movement on smart device screen to restrict the distance. To achieve a fully-automated system, I explore acoustic communications and propose a novel near field authentication system.

Large-scale $\ell_1$-regularized loss minimization problems arise in high-dimensional applications such as compressed sensing and high-dimensional supervised learning, including classification and regression problems. In many applications, it remains challenging to apply the sparse learning model to large-scale problems that have massive data samples with high-dimensional features. One popular and promising strategy is to scaling up the optimization problem in parallel. Parallel solvers run multiple cores on a shared memory system or a distributed environment to speed up the computation, while the practical usage is limited by the huge dimension in the feature space and synchronization problems.

In this dissertation, I carry out the research along the direction with particular focuses on scaling up the optimization of sparse learning for supervised and unsupervised learning problems. For the supervised learning, I firstly propose an asynchronous parallel solver to optimize the large-scale sparse learning model in a multithreading environment. Moreover, I propose a distributed framework to conduct the learning process when the dataset is distributed stored among different machines. Then the proposed model is further extended to the studies of risk genetic factors for Alzheimer's Disease (AD) among different research institutions, integrating a group feature selection framework to rank the top risk SNPs for AD. For the unsupervised learning problem, I propose a highly efficient solver, termed Stochastic Coordinate Coding (SCC), scaling up the optimization of dictionary learning and sparse coding problems. The common issue for the medical imaging research is that the longitudinal features of patients among different time points are beneficial to study together. To further improve the dictionary learning model, I propose a multi-task dictionary learning method, learning the different task simultaneously and utilizing shared and individual dictionary to encode both consistent and changing imaging features.

The widespread adoption of computer vision models is often constrained by the issue of domain mismatch. Models that are trained with data belonging to one distribution, perform poorly when tested with data from a different distribution. Variations in vision based data can be attributed to the following reasons, viz., differences in image quality (resolution, brightness, occlusion and color), changes in camera perspective, dissimilar backgrounds and an inherent diversity of the samples themselves. Machine learning techniques like transfer learning are employed to adapt computational models across distributions. Domain adaptation is a special case of transfer learning, where knowledge from a source domain is transferred to a target domain in the form of learned models and efficient feature representations.

The dissertation outlines novel domain adaptation approaches across different feature spaces; (i) a linear Support Vector Machine model for domain alignment; (ii) a nonlinear kernel based approach that embeds domain-aligned data for enhanced classification; (iii) a hierarchical model implemented using deep learning, that estimates domain-aligned hash values for the source and target data, and (iv) a proposal for a feature selection technique to reduce cross-domain disparity. These adaptation procedures are tested and validated across a range of computer vision applications like object classification, facial expression recognition, digit recognition, and activity recognition. The dissertation also provides a unique perspective of domain adaptation literature from the point-of-view of linear, nonlinear and hierarchical feature spaces. The dissertation concludes with a discussion on the future directions for research that highlight the role of domain adaptation in an era of rapid advancements in artificial intelligence.

Major Depression, clinically called Major Depressive Disorder, is a mood disorder that affects about one eighth of population in US and is projected to be the second leading cause of disability in the world by the year 2020. Recent advances in biotechnology have enabled us to collect a great variety of data which could potentially offer us a deeper understanding of the disorder as well as advancing personalized medicine.

This dissertation focuses on developing methods for three different aspects of predictive analytics related to the disorder: automatic diagnosis, prognosis, and prediction of long-term treatment outcome. The data used for each task have their specific characteristics and demonstrate unique problems. Automatic diagnosis of melancholic depression is made on the basis of metabolic profiles and micro-array gene expression profiles where the presence of missing values and strong empirical correlation between the variables is not unusual. To deal with these problems, a method of generating a representative set of features is proposed. Prognosis is made on data collected from rating scales and questionnaires which consist mainly of categorical and ordinal variables and thus favor decision tree based predictive models. Decision tree models are known for the notorious problem of overfitting. A decision tree pruning method that overcomes the shortcomings of a greedy nature and reliance on heuristics inherent in traditional decision tree pruning approaches is proposed. The method is further extended to prune Gradient Boosting Decision Tree and tested on the task of prognosis of treatment outcome. Follow-up studies evaluating the long-term effect of the treatments on patients usually measure patients' depressive symptom severity monthly, resulting in the actual time of relapse upper bounded by the observed time of relapse. To resolve such uncertainty in response, a general loss function where the hypothesis could take different forms is proposed to predict the risk of relapse in situations where only an interval for time of relapse can be derived from the observed data.

Alzheimer's Disease (AD) is the most common form of dementia observed in elderly patients and has significant social-economic impact. There are many initiatives which aim to capture leading causes of AD. Several genetic, imaging, and biochemical markers are being explored to monitor progression of AD and explore treatment and detection options. The primary focus of this thesis is to identify key biomarkers to understand the pathogenesis and prognosis of Alzheimer's Disease. Feature selection is the process of finding a subset of relevant features to develop efficient and robust learning models. It is an active research topic in diverse areas such as computer vision, bioinformatics, information retrieval, chemical informatics, and computational finance. In this work, state of the art feature selection algorithms, such as Student's t-test, Relief-F, Information Gain, Gini Index, Chi-Square, Fisher Kernel Score, Kruskal-Wallis, Minimum Redundancy Maximum Relevance, and Sparse Logistic regression with Stability Selection have been extensively exploited to identify informative features for AD using data from Alzheimer's Disease Neuroimaging Initiative (ADNI). An integrative approach which uses blood plasma protein, Magnetic Resonance Imaging, and psychometric assessment scores biomarkers has been explored. This work also analyzes the techniques to handle unbalanced data and evaluate the efficacy of sampling techniques. Performance of feature selection algorithm is evaluated using the relevance of derived features and the predictive power of the algorithm using Random Forest and Support Vector Machine classifiers. Performance metrics such as Accuracy, Sensitivity and Specificity, and area under the Receiver Operating Characteristic curve (AUC) have been used for evaluation. The feature selection algorithms best suited to analyze AD proteomics data have been proposed. The key biomarkers distinguishing healthy and AD patients, Mild Cognitive Impairment (MCI) converters and non-converters, and healthy and MCI patients have been identified.

Despite significant advances in digital pathology and automation sciences, current diagnostic practice for cancer detection primarily relies on a qualitative manual inspection of tissue architecture and cell and nuclear morphology in stained biopsies using low-magnification, two-dimensional (2D) brightfield microscopy. The efficacy of this process is limited by inter-operator variations in sample preparation and imaging, and by inter-observer variability in assessment. Over the past few decades, the predictive value quantitative morphology measurements derived from computerized analysis of micrographs has been compromised by the inability of 2D microscopy to capture information in the third dimension, and by the anisotropic spatial resolution inherent to conventional microscopy techniques that generate volumetric images by stacking 2D optical sections to approximate 3D. To gain insight into the analytical 3D nature of cells, this dissertation explores the application of a new technology for single-cell optical computed tomography (optical cell CT) that is a promising 3D tomographic imaging technique which uses visible light absorption to image stained cells individually with sub-micron, isotropic spatial resolution. This dissertation provides a scalable analytical framework to perform fully-automated 3D morphological analysis from transmission-mode optical cell CT images of hematoxylin-stained cells. The developed framework performs rapid and accurate quantification of 3D cell and nuclear morphology, facilitates assessment of morphological heterogeneity, and generates shape- and texture-based biosignatures predictive of the cell state. Custom 3D image segmentation methods were developed to precisely delineate volumes of interest (VOIs) from reconstructed cell images. Comparison with user-defined ground truth assessments yielded an average agreement (DICE coefficient) of 94% for the cell and its nucleus. Seventy nine biologically relevant morphological descriptors (features) were computed from the segmented VOIs, and statistical classification methods were implemented to determine the subset of features that best predicted cell health. The efficacy of our proposed framework was demonstrated on an in vitro model of multistep carcinogenesis in human Barrett's esophagus (BE) and classifier performance using our 3D morphometric analysis was compared against computerized analysis of 2D image slices that reflected conventional cytological observation. Our results enable sensitive and specific nuclear grade classification for early cancer diagnosis and underline the value of the approach as an objective adjunctive tool to better understand morphological changes associated with malignant transformation.

In recent years, machine learning and data mining technologies have received growing attention in several areas such as recommendation systems, natural language processing, speech and handwriting recognition, image processing and biomedical domain. Many of these applications which deal with physiological and biomedical data require person specific or person adaptive systems. The greatest challenge in developing such systems is the subject-dependent data variations or subject-based variability in physiological and biomedical data, which leads to difference in data distributions making the task of modeling these data, using traditional machine learning algorithms, complex and challenging. As a result, despite the wide application of machine learning, efficient deployment of its principles to model real-world data is still a challenge. This dissertation addresses the problem of subject based variability in physiological and biomedical data and proposes person adaptive prediction models based on novel transfer and active learning algorithms, an emerging field in machine learning. One of the significant contributions of this dissertation is a person adaptive method, for early detection of muscle fatigue using Surface Electromyogram signals, based on a new multi-source transfer learning algorithm. This dissertation also proposes a subject-independent algorithm for grading the progression of muscle fatigue from 0 to 1 level in a test subject, during isometric or dynamic contractions, at real-time. Besides subject based variability, biomedical image data also varies due to variations in their imaging techniques, leading to distribution differences between the image databases. Hence a classifier learned on one database may perform poorly on the other database. Another significant contribution of this dissertation has been the design and development of an efficient biomedical image data annotation framework, based on a novel combination of transfer learning and a new batch-mode active learning method, capable of addressing the distribution differences across databases. The methodologies developed in this dissertation are relevant and applicable to a large set of computing problems where there is a high variation of data between subjects or sources, such as face detection, pose detection and speech recognition. From a broader perspective, these frameworks can be viewed as a first step towards design of automated adaptive systems for real world data.