Optimization and Applications of Large-Scale Biomedical Event Networks

The overwhelming amount and unprecedented speed of publication in the biomedical domain make it difficult for life science researchers to acquire and maintain a broad view of the field and gather all information that would be relevant for their research. As a response to this problem, the BioNLP (Biomedical Natural Language Processing) community of researches has emerged and strives to assist life science researchers by developing modern natural language processing (NLP), information extraction (IE) and information retrieval (IR) methods that can be applied at large-scale, to scan the whole publicly available biomedical literature and extract and aggregate the information found within, while automatically normalizing the variability of natural language statements. Among different tasks, biomedical event extraction has received much attention within BioNLP community recently. Biomedical event extraction constitutes the identification of biological processes and interactions described in biomedical literature, and their representation as a set of recursive event structures. The 2009–2013 series of BioNLP Shared Tasks on Event Extraction have given raise to a number of event extraction systems, several of which have been applied at a large scale (the full set of PubMed abstracts and PubMed Central Open Access full text articles), leading to creation of massive biomedical event databases, each of which containing millions of events. Sinece top-ranking event extraction systems are based on machine-learning approach and are trained on the narrow-domain, carefully selected Shared Task training data, their performance drops when being faced with the topically highly varied PubMed and PubMed Central documents. Specifically, false-positive predictions by these systems lead to generation of incorrect biomolecular events which are spotted by the end-users. This thesis proposes a novel post-processing approach, utilizing a combination of supervised and unsupervised learning techniques, that can automatically identify and filter out a considerable proportion of incorrect events from large-scale event databases, thus increasing the general credibility of those databases. The second part of this thesis is dedicated to a system we developed for hypothesis generation from large-scale event databases, which is able to discover novel biomolecular interactions among genes/gene-products. We cast the hypothesis generation problem as a supervised network topology prediction, i.e predicting new edges in the network, as well as types and directions for these edges, utilizing a set of features that can be extracted from large biomedical event networks. Routine machine learning evaluation results, as well as manual evaluation results suggest that the problem is indeed learnable. This work won the Best Paper Award in The 5th International Symposium on Languages in Biology and Medicine (LBM 2013).

Federated architecture for biomedical data integration

The last decades have been characterized by a continuous adoption of IT solutions in the healthcare sector, which resulted in the proliferation of tremendous amounts of data over heterogeneous systems. Distinct data types are currently generated, manipulated, and stored, in the several institutions where patients are treated. The data sharing and an integrated access to this information will allow extracting relevant knowledge that can lead to better diagnostics and treatments. This thesis proposes new integration models for gathering information and extracting knowledge from multiple and heterogeneous biomedical sources. The scenario complexity led us to split the integration problem according to the data type and to the usage specificity. The first contribution is a cloud-based architecture for exchanging medical imaging services. It offers a simplified registration mechanism for providers and services, promotes remote data access, and facilitates the integration of distributed data sources. Moreover, it is compliant with international standards, ensuring the platform interoperability with current medical imaging devices. The second proposal is a sensor-based architecture for integration of electronic health records. It follows a federated integration model and aims to provide a scalable solution to search and retrieve data from multiple information systems. The last contribution is an open architecture for gathering patient-level data from disperse and heterogeneous databases. All the proposed solutions were deployed and validated in real world use cases.

Measuring, Modeling, and Forecasting Volatility and Correlations from High-Frequency Data

This dissertation contains four essays that all share a common purpose: developing new methodologies to exploit the potential of high-frequency data for the measurement, modeling and forecasting of financial assets volatility and correlations. The first two chapters provide useful tools for univariate applications while the last two chapters develop multivariate methodologies. In chapter 1, we introduce a new class of univariate volatility models named FloGARCH models. FloGARCH models provide a parsimonious joint model for low frequency returns and realized measures, and are sufficiently flexible to capture long memory as well as asymmetries related to leverage effects. We analyze the performances of the models in a realistic numerical study and on the basis of a data set composed of 65 equities. Using more than 10 years of high-frequency transactions, we document significant statistical gains related to the FloGARCH models in terms of in-sample fit, out-of-sample fit and forecasting accuracy compared to classical and Realized GARCH models. In chapter 2, using 12 years of high-frequency transactions for 55 U.S. stocks, we argue that combining low-frequency exogenous economic indicators with high-frequency financial data improves the ability of conditionally heteroskedastic models to forecast the volatility of returns, their full multi-step ahead conditional distribution and the multi-period Value-at-Risk. Using a refined version of the Realized LGARCH model allowing for time-varying intercept and implemented with realized kernels, we document that nominal corporate profits and term spreads have strong long-run predictive ability and generate accurate risk measures forecasts over long-horizon. The results are based on several loss functions and tests, including the Model Confidence Set. Chapter 3 is a joint work with David Veredas. We study the class of disentangled realized estimators for the integrated covariance matrix of Brownian semimartingales with finite activity jumps. These estimators separate correlations and volatilities. We analyze different combinations of quantile- and median-based realized volatilities, and four estimators of realized correlations with three synchronization schemes. Their finite sample properties are studied under four data generating processes, in presence, or not, of microstructure noise, and under synchronous and asynchronous trading. The main finding is that the pre-averaged version of disentangled estimators based on Gaussian ranks (for the correlations) and median deviations (for the volatilities) provide a precise, computationally efficient, and easy alternative to measure integrated covariances on the basis of noisy and asynchronous prices. Along these lines, a minimum variance portfolio application shows the superiority of this disentangled realized estimator in terms of numerous performance metrics. Chapter 4 is co-authored with Niels S. Hansen, Asger Lunde and Kasper V. Olesen, all affiliated with CREATES at Aarhus University. We propose to use the Realized Beta GARCH model to exploit the potential of high-frequency data in commodity markets. The model produces high quality forecasts of pairwise correlations between commodities which can be used to construct a composite covariance matrix. We evaluate the quality of this matrix in a portfolio context and compare it to models used in the industry. We demonstrate significant economic gains in a realistic setting including short selling constraints and transaction costs.

RDA and Linked Data: A New Way to Look at the Henson Collection

In 2005, the University of Maryland acquired over 70 digital videos spanning 35 years of Jim Henson’s groundbreaking work in television and film. To support in-house discovery and use, the collection was cataloged in detail using AACR2 and MARC21, and a web-based finding aid was also created. In the past year, I created an "r-ball" (a linked data set described using RDA) of these same resources. The presentation will compare and contrast these three ways of accessing the Jim Henson Works collection, with insights gleaned from providing resource discovery using RIMMF (RDA in Many Metadata Formats).

INVISIBLE LABOR FOR DATA: INSTITUTIONS, INFRASTRUCTURE, AND VIRTUAL SPACE

Americans are accustomed to a wide range of data collection in their lives: census, polls, surveys, user registrations, and disclosure forms. When logging onto the Internet, users’ actions are being tracked everywhere: clicking, typing, tapping, swiping, searching, and placing orders. All of this data is stored to create data-driven profiles of each user. Social network sites, furthermore, set the voluntarily sharing of personal data as the default mode of engagement. But people’s time and energy devoted to creating this massive amount of data, on paper and online, are taken for granted. Few people would consider their time and energy spent on data production as labor. Even if some people do acknowledge their labor for data, they believe it is accessory to the activities at hand. In the face of pervasive data collection and the rising time spent on screens, why do people keep ignoring their labor for data? How has labor for data been become invisible, as something that is disregarded by many users? What does invisible labor for data imply for everyday cultural practices in the United States? Invisible Labor for Data addresses these questions. I argue that three intertwined forces contribute to framing data production as being void of labor: data production institutions throughout history, the Internet’s technological infrastructure (especially with the implementation of algorithms), and the multiplication of virtual spaces. There is a common tendency in the framework of human interactions with computers to deprive data and bodies of their materiality. My Introduction and Chapter 1 offer theoretical interventions by reinstating embodied materiality and redefining labor for data as an ongoing process. The middle Chapters present case studies explaining how labor for data is pushed to the margin of the narratives about data production. I focus on a nationwide debate in the 1960s on whether the U.S. should build a databank, contemporary Big Data practices in the data broker and the Internet industries, and the group of people who are hired to produce data for other people’s avatars in the virtual games. I conclude with a discussion on how the new development of crowdsourcing projects may usher in the new chapter in exploiting invisible and discounted labor for data.

BUILDING EFFICIENT AND COST-EFFECTIVE CLOUD-BASED BIG DATA MANAGEMENT SYSTEMS

In today’s big data world, data is being produced in massive volumes, at great velocity and from a variety of different sources such as mobile devices, sensors, a plethora of small devices hooked to the internet (Internet of Things), social networks, communication networks and many others. Interactive querying and large-scale analytics are being increasingly used to derive value out of this big data. A large portion of this data is being stored and processed in the Cloud due the several advantages provided by the Cloud such as scalability, elasticity, availability, low cost of ownership and the overall economies of scale. There is thus, a growing need for large-scale cloud-based data management systems that can support real-time ingest, storage and processing of large volumes of heterogeneous data. However, in the pay-as-you-go Cloud environment, the cost of analytics can grow linearly with the time and resources required. Reducing the cost of data analytics in the Cloud thus remains a primary challenge. In my dissertation research, I have focused on building efficient and cost-effective cloud-based data management systems for different application domains that are predominant in cloud computing environments. In the first part of my dissertation, I address the problem of reducing the cost of transactional workloads on relational databases to support database-as-a-service in the Cloud. The primary challenges in supporting such workloads include choosing how to partition the data across a large number of machines, minimizing the number of distributed transactions, providing high data availability, and tolerating failures gracefully. I have designed, built and evaluated SWORD, an end-to-end scalable online transaction processing system, that utilizes workload-aware data placement and replication to minimize the number of distributed transactions that incorporates a suite of novel techniques to significantly reduce the overheads incurred both during the initial placement of data, and during query execution at runtime. In the second part of my dissertation, I focus on sampling-based progressive analytics as a means to reduce the cost of data analytics in the relational domain. Sampling has been traditionally used by data scientists to get progressive answers to complex analytical tasks over large volumes of data. Typically, this involves manually extracting samples of increasing data size (progressive samples) for exploratory querying. This provides the data scientists with user control, repeatable semantics, and result provenance. However, such solutions result in tedious workflows that preclude the reuse of work across samples. On the other hand, existing approximate query processing systems report early results, but do not offer the above benefits for complex ad-hoc queries. I propose a new progressive data-parallel computation framework, NOW!, that provides support for progressive analytics over big data. In particular, NOW! enables progressive relational (SQL) query support in the Cloud using unique progress semantics that allow efficient and deterministic query processing over samples providing meaningful early results and provenance to data scientists. NOW! enables the provision of early results using significantly fewer resources thereby enabling a substantial reduction in the cost incurred during such analytics. Finally, I propose NSCALE, a system for efficient and cost-effective complex analytics on large-scale graph-structured data in the Cloud. The system is based on the key observation that a wide range of complex analysis tasks over graph data require processing and reasoning about a large number of multi-hop neighborhoods or subgraphs in the graph; examples include ego network analysis, motif counting in biological networks, finding social circles in social networks, personalized recommendations, link prediction, etc. These tasks are not well served by existing vertex-centric graph processing frameworks whose computation and execution models limit the user program to directly access the state of a single vertex, resulting in high execution overheads. Further, the lack of support for extracting the relevant portions of the graph that are of interest to an analysis task and loading it onto distributed memory leads to poor scalability. NSCALE allows users to write programs at the level of neighborhoods or subgraphs rather than at the level of vertices, and to declaratively specify the subgraphs of interest. It enables the efficient distributed execution of these neighborhood-centric complex analysis tasks over largescale graphs, while minimizing resource consumption and communication cost, thereby substantially reducing the overall cost of graph data analytics in the Cloud. The results of our extensive experimental evaluation of these prototypes with several real-world data sets and applications validate the effectiveness of our techniques which provide orders-of-magnitude reductions in the overheads of distributed data querying and analysis in the Cloud.

Rational Function Interpolation of Electromagnetic Transfer Functions of High-Speed Interconnect Systems from Discrete Time-Domain and Frequency-Domain Data

We present new methodologies to generate rational function approximations of broadband electromagnetic responses of linear and passive networks of high-speed interconnects, and to construct SPICE-compatible, equivalent circuit representations of the generated rational functions. These new methodologies are driven by the desire to improve the computational efficiency of the rational function fitting process, and to ensure enhanced accuracy of the generated rational function interpolation and its equivalent circuit representation. Toward this goal, we propose two new methodologies for rational function approximation of high-speed interconnect network responses. The first one relies on the use of both time-domain and frequency-domain data, obtained either through measurement or numerical simulation, to generate a rational function representation that extrapolates the input, early-time transient response data to late-time response while at the same time providing a means to both interpolate and extrapolate the used frequency-domain data. The aforementioned hybrid methodology can be considered as a generalization of the frequency-domain rational function fitting utilizing frequency-domain response data only, and the time-domain rational function fitting utilizing transient response data only. In this context, a guideline is proposed for estimating the order of the rational function approximation from transient data. The availability of such an estimate expedites the time-domain rational function fitting process. The second approach relies on the extraction of the delay associated with causal electromagnetic responses of interconnect systems to provide for a more stable rational function process utilizing a lower-order rational function interpolation. A distinctive feature of the proposed methodology is its utilization of scattering parameters. For both methodologies, the approach of fitting the electromagnetic network matrix one element at a time is applied. It is shown that, with regard to the computational cost of the rational function fitting process, such an element-by-element rational function fitting is more advantageous than full matrix fitting for systems with a large number of ports. Despite the disadvantage that different sets of poles are used in the rational function of different elements in the network matrix, such an approach provides for improved accuracy in the fitting of network matrices of systems characterized by both strongly coupled and weakly coupled ports. Finally, in order to provide a means for enforcing passivity in the adopted element-by-element rational function fitting approach, the methodology for passivity enforcement via quadratic programming is modified appropriately for this purpose and demonstrated in the context of element-by-element rational function fitting of the admittance matrix of an electromagnetic multiport.

Detection of mesoscale thermal fronts from 4km data using smoothing techniques: Gradient-based fronts classification and basin scale application

In order to optimize frontal detection in sea surface temperature fields at 4 km resolution, a combined statistical and expert-based approach is applied to test different spatial smoothing of the data prior to the detection process. Fronts are usually detected at 1 km resolution using the histogram-based, single image edge detection (SIED) algorithm developed by Cayula and Cornillon in 1992, with a standard preliminary smoothing using a median filter and a 3 × 3 pixel kernel. Here, detections are performed in three study regions (off Morocco, the Mozambique Channel, and north-western Australia) and across the Indian Ocean basin using the combination of multiple windows (CMW) method developed by Nieto, Demarcq and McClatchie in 2012 which improves on the original Cayula and Cornillon algorithm. Detections at 4 km and 1 km of resolution are compared. Fronts are divided in two intensity classes (“weak” and “strong”) according to their thermal gradient. A preliminary smoothing is applied prior to the detection using different convolutions: three type of filters (median, average and Gaussian) combined with four kernel sizes (3 × 3, 5 × 5, 7 × 7, and 9 × 9 pixels) and three detection window sizes (16 × 16, 24 × 24 and 32 × 32 pixels) to test the effect of these smoothing combinations on reducing the background noise of the data and therefore on improving the frontal detection. The performance of the combinations on 4 km data are evaluated using two criteria: detection efficiency and front length. We find that the optimal combination of preliminary smoothing parameters in enhancing detection efficiency and preserving front length includes a median filter, a 16 × 16 pixel window size, and a 5 × 5 pixel kernel for strong fronts and a 7 × 7 pixel kernel for weak fronts. Results show an improvement in detection performance (from largest to smallest window size) of 71% for strong fronts and 120% for weak fronts. Despite the small window used (16 × 16 pixels), the length of the fronts has been preserved relative to that found with 1 km data. This optimal preliminary smoothing and the CMW detection algorithm on 4 km sea surface temperature data are then used to describe the spatial distribution of the monthly frequencies of occurrence for both strong and weak fronts across the Indian Ocean basin. In general strong fronts are observed in coastal areas whereas weak fronts, with some seasonal exceptions, are mainly located in the open ocean. This study shows that adequate noise reduction done by a preliminary smoothing of the data considerably improves the frontal detection efficiency as well as the global quality of the results. Consequently, the use of 4 km data enables frontal detections similar to 1 km data (using a standard median 3 × 3 convolution) in terms of detectability, length and location. This method, using 4 km data is easily applicable to large regions or at the global scale with far less constraints of data manipulation and processing time relative to 1 km data.

Design and implementation of a floating point unit for rigel, a massively parallel accelerator

Scientific applications rely heavily on floating point data types. Floating point operations are complex and require complicated hardware that is both area and power intensive. The emergence of massively parallel architectures like Rigel creates new challenges and poses new questions with respect to floating point support. The massively parallel aspect of Rigel places great emphasis on area efficient, low power designs. At the same time, Rigel is a general purpose accelerator and must provide high performance for a wide class of applications. This thesis presents an analysis of various floating point unit (FPU) components with respect to Rigel, and attempts to present a candidate design of an FPU that balances performance, area, and power and is suitable for massively parallel architectures like Rigel.