Digital Commons Network^™

Developing Agent-Based Models To Study Financial Markets, Saurav Chakraborty

USF Tampa Graduate Theses and Dissertations

This dissertation presents research that employs agent-based modelling to provide a framework to support simulation as a complement to traditional economic models for policy evaluation. It consists of three studies. The first study employs cluster analysis to capture the different types of banks and the associated business models that define their decision-making. The results from study one will help us get an understanding of how different banks behave and provide an insight into their lending practices. Hence, it would be very helpful in evaluating and analyzing the impact of future policies. Study two develops a fine-grained interbank lending model based …

Functional Object-Oriented Network: A Knowledge Representation For Service Robotics, David Andrés Paulius Ramos

USF Tampa Graduate Theses and Dissertations

In this dissertation, we discuss our work behind the development of the functional object-oriented network (abbreviated as FOON), a graphical knowledge representation for robotic manipulation and understanding of its own actions and (potentially) the intentions of humans in the household. Based on the theory of affordance, this representation captures manipulations and their effects on actions through the coupling of object and motion nodes as fundamental learning units known as functional units. The activities currently represented in FOON are cooking related, but this representation can be extended to other activities that involve manipulation of objects which result in observable changes of …

Toward Culturally Relevant Emotion Detection Using Physiological Signals, Khadija Zanna

USF Tampa Graduate Theses and Dissertations

Research shows that emotional distress has a statistically significant impact on a student’s grade point average and intent to drop out of college. Because students of different races have varying college experiences, it is important to understand the emotional experiences of different racial groups to better support students’ needs and academic success. In this work, we explore several physiological responses to ten different emotional stimuli captured from 140 students. We employ unsupervised learning via the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and supervised learning via Random Forests and Support Vector machines to analyze clustering partitions and classification …

Algorithms To Profile Driver Behavior From Zero-Permission Embedded Sensors, Bharti Goel

USF Tampa Graduate Theses and Dissertations

In this dissertation, we design algorithms to profile driver behavior from zero-permission sensors embedded in modern smartphones and wearables. These sensors are typically the accelerometer, gyroscope, magnetometer, pressure sensor and a few more than are now available in most modern smartphones and wearables. In order to profile driving behavior, we devised algorithms for detecting distraction while driving due to the use of modern-day smartphones (e.g., calling, texting and reading while driving) in real-time.

To do so, we conduct an experiment with 16 subjects on a realistic driving simulator, where each subject, where each subject carries a smartphone and a wearable …

Reliability Analysis Of Power Grids And Its Interdependent Infrastructures: An Interaction Graph-Based Approach, Upama Nakarmi

USF Tampa Graduate Theses and Dissertations

Large blackouts with significant societal and economic impacts result from cascade of failures in the transmission network of power grids. Understanding and mitigating cascading failures in power grids is challenging due to the large number of components and their complex interactions, wherein, in addition to the physical topology of the system, the physics of power flow and functional dependencies among components largely affect the spatial distribution and propagation of failures. In this dissertation, data-driven interaction graphs, which help in capturing the underlying interactions and influences among the components during cascading failures, are used for capturing the non-local nature of propagation …

Towards Safe Power Oversubscription And Energy Efficiency Of Data Centers, Sulav Malla

USF Tampa Graduate Theses and Dissertations

Data centers contribute to approximately 1% of the global electricity consumption, and billions of dollars are spent annually worldwide in construction of new data centers to meet the rising demand for cloud-based services. Given the high cost of construction, the power infrastructure in a data center is typically oversubscribed. Power oversubscription leads to efficient use of data center power hierarchy while simultaneously reducing the power provisioning cost. Power overload situations can occur in oversubscribed data centers. Power overload can lead to power capping of servers or even power outages – both of which degrade the performance of the services offered …

Lung Nodule Malignancy Prediction From Computed Tomography Images Using Deep Learning, Rahul Paul

USF Tampa Graduate Theses and Dissertations

Lung cancer has a high incidence and mortality rate. The five-year relative survival rate for all lung cancers is 18%. Due to the high mortality and incidence rate of lung cancer worldwide, early detection is essential. Low dose Computed Tomography (CT) is a commonly used technique for screening, diagnosis, and prognosis of non-small cell lung cancer (NSCLC). The National Lung Screening Trial (NLST) compared low-dose helical computed tomography (LDCT) and standard chest radiography (CXR) for three annual screens and reported a 20% relative reduction in lung cancer mortality for LDCT compared to CXR. As such, LDCT screening for lung cancer …

Algorithms And Framework For Computing 2-Body Statistics On Graphics Processing Units, Napath Pitaksirianan

USF Tampa Graduate Theses and Dissertations

Various types of two-body statistics (2-BS) are regarded as essential components of low-level data analysis in scientific database systems. In relational algebraic terms, a 2-BS is essentially a Cartesian product between two datasets (or two instances of the same dataset) followed by a user-defined aggregate. The quadratic complexity of these computations hinders the timely processing of data. Thus using modern parallel hardware has become an obvious solution to meet such challenges. This dissertation presents our recent work in designing and optimizing parallel algorithms for 2-BS computation on Graphics Processing Units (GPUs). The unique architecture, however, provides abundant opportunities for optimizing …

Efficient Forward-Secure And Compact Signatures For The Internet Of Things (Iot), Efe Ulas Akay Seyitoglu

USF Tampa Graduate Theses and Dissertations

In the modern Internet of Things (IoT) applications, the system entities collect security-sensitive information that must be cryptographically protected. In particular, authentication and integrity, as foundational security services, are essential for any IoT applications. Digital signatures provide both authentication and integrity to these applications. Nevertheless, once an IoT device is compromised, its signature private key is leaked to an adversary. Forward-secure digital signatures mitigate the impact of such key compromises by incorporating a key-evolving mechanism into the authentication process. However, existing forward-secure signatures suffer from large signature/key sizes, heavy computational overhead, and some prominent variants that can only sign a …

Psidb: A Framework For Batched Query Processing And Optimization, Mehrad Eslami

USF Tampa Graduate Theses and Dissertations

Techniques based on sharing data and computation among queries have been an active research topic in database systems. While work in this area developed algorithms and systems that are shown to be effective, there is a lack of logical foundation for query processing and optimization. In this paper, we present PsiDB, a system model for processing a large number of database queries in a batch. The key idea is to generate a single query expression that returns a global relation containing all the data needed for individual queries. For that, we propose the use of a type of relational operators …

Classifying Emotions With Eeg And Peripheral Physiological Data Using 1d Convolutional Long Short-Term Memory Neural Network, Rupal Agarwal

USF Tampa Graduate Theses and Dissertations

Recognizing emotions is very important while building robust and interactive Affective Brain-Computer Interfaces as it allows the machines to have some degree of emotional intelligence with the help of which they can understand the changing emotional state of users. In the past, emotions have been recognized via unimodal data such as electroencephalography (EEG) signals, speech, facial expressions or peripheral physiological signals. However, emotions are complex as they are a combination of human behavior, thinking and feeling. Therefore, as compared to unimodal methods, multi-modal techniques, recognize emotions with more reliability. This thesis aims to recognize and classify human emotions into high/low …