University of North Texas

About: University of North Texas is a education organization based out in Denton, Texas, United States. It is known for research contribution in the topics: Population & Poison control. The organization has 11866 authors who have published 26984 publications receiving 705376 citations. The organization is also known as: Fight, North Texas & UNT.

Three Decades of Research on Motivational Intensity Theory: What We Have Learned About Effort and What We Still Don't Know

Abstract: Brehm's motivational intensity theory has been a fruitful conceptual framework for research on effort during the last three decades. Researchers have used the theory to address various effort-related phenomena, like the impact of ability, affect, and fatigue on effort mobilization. In this chapter, we provide an overview of development in the last 10 years focusing on research that has addressed (1) the energy conservation principle, (2) ability and fatigue effects, and (3) the impact of mood, dysphoria, and primed affect. We point out that most of the research has supported the predictions of the theory and its extensions and applications. However, we also elaborate on empirical findings that do not fit the theory and discuss open questions that need to be addressed in future research.

Learning Sentiment Lexicons in Spanish

Abstract: In this paper we present a framework to derive sentiment lexicons in a target language by using manually or automatically annotated data available in an electronic resource rich language, such as English. We show that bridging the language gap using the multilingual sense-level aligned WordNet structure allows us to generate a high accuracy (90%) polarity lexicon comprising 1,347 entries, and a disjoint lower accuracy (74%) one encompassing 2,496 words. By using an LSA-based vectorial expansion for the generated lexicons, we are able to obtain an average F-measure of 66% in the target language. This implies that the lexicons could be used to bootstrap higher-coverage lexicons using in-language resources.

Privacy-preserving Machine Learning as a Service

Abstract: Machine learning algorithms based on neural networks have achieved remarkable results and are being extensively used in different domains. However, the machine learning algorithms requires access to raw data which is often privacy sensitive. To address this issue, we develop new techniques to provide solutions for running deep neural networks over encrypted data. In this paper, we develop new techniques to adopt deep neural networks within the practical limitation of current homomorphic encryption schemes. We focus on training and classification of the well-known neural networks and convolutional neural networks. First, we design methods for approximation of the activation functions commonly used in CNNs (i.e. ReLU, Sigmoid, and Tanh) with low degree polynomials which is essential for efficient homomorphic encryption schemes. Then, we train neural networks with the approximation polynomials instead of original activation functions and analyze the performance of the models. Finally, we implement neural networks and convolutional neural networks over encrypted data and measure performance of the models.

Rationally designed graphene-nanotube 3D architectures with a seamless nodal junction for efficient energy conversion and storage.

Abstract: One-dimensional (1D) carbon nanotubes (CNTs) and 2D single-atomic layer graphene have superior thermal, electrical, and mechanical properties. However, these nanomaterials exhibit poor out-of-plane properties due to the weak van der Waals interaction in the transverse direction between graphitic layers. Recent theoretical studies indicate that rationally designed 3D architectures could have desirable out-of-plane properties while maintaining in-plane properties by growing CNTs and graphene into 3D architectures with a seamless nodal junction. However, the experimental realization of seamlessly-bonded architectures remains a challenge. We developed a strategy of creating 3D graphene-CNT hollow fibers with radially aligned CNTs (RACNTs) seamlessly sheathed by a cylindrical graphene layer through a one-step chemical vapor deposition using an anodized aluminum wire template. By controlling the aluminum wire diameter and anodization time, the length of the RACNTs and diameter of the graphene hollow fiber can be tuned, enabling efficient energy conversion and storage. These fibers, with a controllable surface area, meso-/micropores, and superior electrical properties, are excellent electrode materials for all-solid-state wire-shaped supercapacitors with poly(vinyl alcohol)/H2SO4 as the electrolyte and binder, exhibiting a surface-specific capacitance of 89.4 mF/cm2 and length-specific capacitance up to 23.9 mF/cm, — one to four times the corresponding record-high capacities reported for other fiber-like supercapacitors. Dye-sensitized solar cells, fabricated using the fiber as a counter electrode, showed a power conversion efficiency of 6.8% and outperformed their counterparts with an expensive Pt wire counter electrode by a factor of 2.5. These novel fiber-shaped graphene-RACNT energy conversion and storage devices are so flexible they can be woven into fabrics as power sources.