Physical Sciences and Mathematics Commons^™

Geometric Unified Method In 3d Object Classification, Mengyi Shan

HMC Senior Theses

3D object classification is one of the most popular topics in the field of computer vision and computational geometry. Currently, the most popular state-of-the-art algorithm is the so-called Convolutional Neural Network (CNN) models with various representations that capture different features of the given 3D data, including voxels, local features, multi-view 2D features, and so on. With CNN as a holistic approach, researches focus on improving the accuracy and efficiency by designing the neural network architecture. This thesis aims to examine the existing work on 3D object classification and explore the underlying theory by integrating geometric approaches. By using geometric algorithms ...

Radial Singular Solutions To Semilinear Partial Differential Equations, Marcelo A. Almora Rios

HMC Senior Theses

We show the existence of countably many non-degenerate continua of singular radial solutions to a p-subcritical, p-Laplacian Dirichlet problem on the unit ball in R^N. This result generalizes those for the 2-Laplacian to any value p and extends recent work on the p-Laplacian by considering solutions both radial and singular.

An Enticing Study Of Prime Numbers Of The Shape 𝑝 = 𝑥^2 + 𝑦^2, Xiaona Zhou

Publications and Research

We will study and prove important results on primes of the shape 𝑥² + 𝑦² using number theoretic techniques. Our analysis involves maps, actions over sets, fixed points and involutions. This presentation is readily accessible to an advanced undergraduate student and lay the groundwork for future studies.

A Near-Optimal Change-Detection Based Algorithm For Piecewise-Stationary Combinatorial Semi-Bandits, Huozhi Zhou, Lingda Wang, Lav N. Varshney, Ee Peng Lim

Research Collection School Of Information Systems

We investigate the piecewise-stationary combinatorial semi-bandit problem. Compared to the original combinatorial semi-bandit problem, our setting assumes the reward distributions of base arms may change in a piecewise-stationary manner at unknown time steps. We propose an algorithm, GLR-CUCB, which incorporates an efficient combinatorial semi-bandit algorithm, CUCB, with an almost parameter-free change-point detector, the Generalized Likelihood Ratio Test (GLRT). Our analysis shows that the regret of GLR-CUCB is upper bounded by O(√NKT logT), where N is the number of piecewise-stationary segments, K is the number of base arms, and T is the number of time steps. As a complement, we ...

Corn And Soybean Response To Wastewater-Recycled Phosphorus Fertilizers, Shane Ylagan

Crop, Soil and Environmental Sciences Undergraduate Honors Theses

The ability to recycle phosphorus (P) from wastewaters could provide a sustainable, continuous source of P that might also help protect surface water quality from P enrichment. The mineral struvite (MgNH4PO4·6H2O) is an understudied material that can be created from Pcontaining wastewater and has been shown to have agricultural fertilizer value. The objective of this study was to evaluate the effects of electrochemically precipitated struvite (ECST), chemically precipitated struvite (Crystal Green; CG), diammonium phosphate (DAP), monoammonium phosphate (MAP), rock phosphate (RP), and triple super phosphate (TSP) on corn (Zea mays) and soybean (Glycine max) response in a 79-day greenhouse ...

Teaching Applications And Implications Of Blockchain Via Project-Based Learning: A Case Study, Kevin Mentzer, Mark Frydenberg, David J. Yates

Computer Information Systems Journal Articles

This paper presents student projects analyzing or using blockchain technologies, created by students enrolled in courses dedicated to teaching blockchain, at two different universities during the 2018-2019 academic year. Students explored perceptions related to storing private healthcare information on a blockchain, managing the security of Internet of Things devices, maintaining public governmental records, and creating smart contracts. The course designs, which were centered around project-based learning, include self-regulated learning and peer feedback as ways to improve student learning. Students either wrote a research paper or worked in teams on a programming project to build and deploy a blockchain-based application using ...

Detecting Hacker Threats: Performance Of Word And Sentence Embedding Models In Identifying Hacker Communications, Susan Mckeever, Brian Keegan, Andrei Quieroz

Conference papers

Abstract—Cyber security is striving to find new forms of protection against hacker attacks. An emerging approach nowadays is the investigation of security-related messages exchanged on deep/dark web and even surface web channels. This approach can be supported by the use of supervised machine learning models and text mining techniques. In our work, we compare a variety of machine learning algorithms, text representations and dimension reduction approaches for the detection accuracies of software-vulnerability-related communications. Given the imbalanced nature of the three public datasets used, we investigate appropriate sampling approaches to boost detection accuracies of our models. In addition, we ...

Approximating Special Social Influence Maximization Problems, Jie Wu, Ning Wang

Faculty Scholarship for the College of Science & Mathematics

Social Influence Maximization Problems (SIMPs) deal with selecting k seeds in a given Online Social Network (OSN) to maximize the number of eventually-influenced users. This is done by using these seeds based on a given set of influence probabilities among neighbors in the OSN. Although the SIMP has been proved to be NP-hard, it has both submodular (with a natural diminishing-return) and monotone (with an increasing influenced users through propagation) that make the problem suitable for approximation solutions. However, several special SIMPs cannot be modeled as submodular or monotone functions. In this paper, we look at several conditions under which ...

Modification Of Chm20 Lab: Synthesis Of Aspirin, Nicole Diaz, Amy Musselman Prof.

KU Inspires! Undergraduate Research Day

No abstract provided.

Issues And Proposed Solutions To Team Based Assessment In University Physics Labs, Manqoba Q. Hlatshwayo

STEM Instructional Graduate Teaching Assistant Posters

Many education experts considers peer-to-peer education as the most effective way for students to learn (Peersdom, 2014). Western Michigan University (WMU) physics laboratories embraces this philosophy and has designed the assessment to be based on the team report. The assumption is that every team member contributed (not necessarily equally) to the work submitted by the team hence every team member should receive the same grade. This is a ubiquitous assessment philosophy in education, and some forms of additional individual assessment in team projects are strongly discouraged (David Boud et al, 1999).

There is no doubt on the effectiveness of peer-to-peer ...

Improving Lab Experience For Students In Phys2080, Khushi Bhatt

STEM Instructional Graduate Teaching Assistant Posters

University Physics II (PHYS 2080) deals with providing hands-on experience with the experiments of electricity and magnetism. It aids students to develop computational thinking in physics through analysis of experimental and theoretical data using python programming language.

In Fall-2019, I taught the newly designed PHYS2080 class for the first time as per the requirements. During Spring-2020, I implemented some subtle changes using the skills that I gained through various teaching training. These changes are described in Methodology section. Data clearly shows improvement that was observed without lowering the expectations or changing the curriculum of the class.

Constructing A Time Scale Of Biotic Recovery Across The Cretaceous–Paleogene Boundary, Corral Bluffs, Denver Basin, Colorado, U.S.A., Vincent Isakson, Mark D. Schmitz

Geosciences Faculty Publications and Presentations

The Cretaceous–Paleogene (K–Pg) boundary interval represents one of the most significant mass extinctions and ensuing biotic recoveries in Earth history. Earliest Paleocene fossil mammal faunas corresponding to the Puercan North American Land Mammal Age (NALMA) are thought to be highly endemic and potentially diachronous, necessitating precise chronostratigraphic controls at key fossil localities to constrain recovery dynamics in continental biotas following the K–Pg mass extinction. The Laramide synorgenic sedimentary deposits within the Denver Basin in east-central Colorado preserve one of the most continuous and fossiliferous records of the K–Pg boundary interval in North America. Poor exposure in ...

Experimental Comparison Of Features And Classifiers For Android Malware Detection, Lwin Khin Shar, Biniam Fisseha Demissie, Mariano Ceccato, Wei Minn

Research Collection School Of Information Systems

Android platform has dominated the smart phone market for years now and, consequently, gained a lot of attention from attackers. Malicious apps (malware) pose a serious threat to the security and privacy of Android smart phone users. Available approaches to detect mobile malware based on machine learning rely on features extracted with static analysis or dynamic analysis techniques. Dif- ferent types of machine learning classi ers (such as support vector machine and random forest) deep learning classi ers (based on deep neural networks) are then trained on extracted features, to produce models that can be used to detect mobile malware ...

Camera Placement Meeting Restrictions Of Computer Vision, Sara Aghajanzadeh, Roopasree Naidu, Shuo-Han Chen, Caleb Tung, Abhinav Goel, Yung-Hsiang Lu, George K. Thiruvathukal

Computer Science: Faculty Publications and Other Works

In the blooming era of smart edge devices, surveillance cam- eras have been deployed in many locations. Surveillance cam- eras are most useful when they are spaced out to maximize coverage of an area. However, deciding where to place cam- eras is an NP-hard problem and researchers have proposed heuristic solutions. Existing work does not consider a signifi- cant restriction of computer vision: in order to track a moving object, the object must occupy enough pixels. The number of pixels depends on many factors (how far away is the object? What is the camera resolution? What is the focal length ...

Substitutional And Interstitial Doping In Laco5 System For The Development Of Hard Magnetic Properties: A First Principles Study, Huseyin Ucar, Renu Choudhary, Durga Paudyal

Ames Laboratory Accepted Manuscripts

We investigate here the changes in the electronic structure at the transition metal sites of the RE-TM5 structure (RE = Rare Earth, TM = Transition Metal) while doping the interstitial sites with nitrogen. LaCo5 compound is taken as the baseline compound owing to its critically needed intrinsic magnetic properties such as magneto-crystalline anisotropy energy (MAE) of ≈5 meV/fu [1] due to the contributions from the cobalt network. In addition, because of the lack of 4f electrons in lanthanum, complications originating from the treatment of the 4f localized electrons are absent in this compound; making it an ideal reference material to all ...

Late-Holocene Shoreline Responses To Competing Shelf, Bay, And Beach Accommodation Spaces Under Conditions Of Relative Sea Level Change, And The Potential For Future Catastrophic Beach Retreat In The Columbia River Littoral Cell, Washington And Oregon, Usa, Curt D. Peterson, Tamara C. Linde, Sandy Vanderburgh

Geology Faculty Publications and Presentations

The Columbia River Littoral Cell (CRLC) (160 km in length) provides opportunities to compare competing accommodation space relations under different conditions of relative sea level change. The CRLC system includes abundant littoral sand supply from the large Columbia River, late-Holocene prograded beach plains and barrier spits (0.5–5 km in width), two large marine-dominated estuaries (Willapa Bay and Grays Harbor), and a high-wave-energy inner-shelf. Littoral sand accumulation rates in prograded beach plains and barrier deposits are based on paleo-shoreline positions that are dated by great-earthquake catastrophic beach retreat scarps (n = 10) from 0.3 to 5.0 ka. The ...

Comparing Uganda's Indigenous Soil Classification System With World Reference Base And Usda Soil Taxonomy To Predict Soil Productivity, Stewart Kyebogola, C. Lee Burras, Bradley A. Miller, Onesimus Semalulu, Russell S. Yost, Moses M. Tenywa, Andrew W. Lenssen, Prossy Kyomuhendo, Christopher Smith, Charles K. Luswata, Mwanjalolo J. Gilbert Majaliwa, Lance Goettsch, Carol J. Pierce Colfer, Robert E. Mazur

Agronomy Publications

This study examines three soil classification systems - Buganda, World Reference Base, and US Soil Taxonomy - in order to evaluate their relative strengths and feasibility for making linkages between them. Nine field sites and 16 pedons were considered across the soil landscapes of the Buganda catena. Each identified field pedon diagnostic horizons and characteristics were described and their soils analyzed using standard pedological techniques and measurements. To document the indigenous use of the Buganda classification system, interviews and discussions were held with farmer groups and local extension specialists. Using this local expertise, five local soil units were identified. We also identified ...

Temporal Heterogeneous Interaction Graph Embedding For Next-Item Recommendation, Yugang Ji, Mingyang Yin, Yuan Fang, Hongxia Yang, Xiangwei Wang, Tianrui Jia, Chuan Shi

Research Collection School Of Information Systems

In the scenario of next-item recommendation, previous methods attempt to model user preferences by capturing the evolution of sequential interactions. However, their sequential expression is often limited, without modeling complex dynamics that short-term demands can often be influenced by long-term habits. Moreover, few of them take into account the heterogeneous types of interaction between users and items. In this paper, we model such complex data as a Temporal Heterogeneous Interaction Graph (THIG) and learn both user and item embeddings on THIGs to address next-item recommendation. The main challenges involve two aspects: the complex dynamics and rich heterogeneity of interactions. We ...

Social Influence Attentive Neural Network For Friend-Enhanced Recommendation, Yuanfu Lu, Ruobing Xie, Chuan Shi, Yuan Fang, Wei Wang, Xu Zhang, Leyu Lin

Research Collection School Of Information Systems

With the thriving of online social networks, there emerges a new recommendation scenario in many social apps, called FriendEnhanced Recommendation (FER) in this paper. In FER, a user is recommended with items liked/shared by his/her friends (called a friend referral circle). These friend referrals are explicitly shown to users. Different from conventional social recommendation, the unique friend referral circle in FER may significantly change the recommendation paradigm, making users to pay more attention to enhanced social factors. In this paper, we first formulate the FER problem, and propose a novel Social Influence Attentive Neural network (SIAN) solution. In ...

Hierarchical Multimodal Attention For End-To-End Audio-Visual Scene-Aware Dialogue Response Generation, Hung Le, Doyen Sahoo, Nancy F. Chen, Steven C. H. Hoi

Research Collection School Of Information Systems

This work is extended from our participation in the Dialogue System Technology Challenge (DSTC7), where we participated in the Audio Visual Scene-aware Dialogue System (AVSD) track. The AVSD track evaluates how dialogue systems understand video scenes and responds to users about the video visual and audio content. We propose a hierarchical attention approach on user queries, video caption, audio and visual features that contribute to improved evaluation results. We also apply a nonlinear feature fusion approach to combine the visual and audio features for better knowledge representation. Our proposed model shows superior performance in terms of both objective evaluation and ...