Six graduate presentations and defenses this week

CSE graduate defenses
CSE graduate defenses

"Impact of Users' Locus of Control on Presence and Performance in Telepresence Robot Operation"

Urja Acharya

Advisor: Brittany Duncan
Committee members: Justin Bradley, Carrick Detweiler
Date: April 5, 2018
Time: 10 a.m.
Room: 211 Schorr Center

Abstract: This thesis presents research investigating the impact of user qualities on presence and performance with the goal of guiding the development of shared autonomy algorithms to adapt to users based on inferred qualities. Previous works in shared autonomy have focused on adapting to a single user in a single interaction or the ability to infer future states but have neglected sensing and adapting in real-time to personal qualities (e.g. locus of control (LOC)) of users. This study collected user commands, performance, and personal data from 60 participants in a telepresence robot driving task to understand their relationship and generate a strategy for shared autonomy systems which adapt to individual users. This work impacts the human-robot interaction community through its expansion of previous findings in the community, and the communities affiliated with robotics and autonomous systems as a whole in order to better adapt to the novice users of the future. This work also stands to tighten integration between the findings of the HRI community and the design of autonomy in systems. The results found that the time taken by “High External” users was 33.89% more and the distance travelled by these users was 27.62% more than that of the “Average” users in restrictive mode. In relaxed mode, presence perceived by “High Internal” users was 4.79% more than that of the “Average” users. It was also found that the “High Internal” users issued 29.12% more commands, had 69.17% more conflicting commands, 33.2% more percentage of command conflicts, took 41.18% longer duration, and travelled 17.74% farther in restrictive mode in comparison to relaxed mode. These results indicate that users with different LOC performed differently in different modes of shared control and users seeking more control fought against autonomy in restrictive mode but performed better in relaxed mode. These findings suggest that user qualities can be inferred from a brief set of interactions and autonomy can potentially be adapted during runtime to improve user performance.




"Modeling and Visualizing Molecular Information in Metabolic Control: a Multi-layer Perspective"

Aditya Immaneni

Advisor: Massimiliano Pierobon
Committee members: Juan Cui, Tomas Hellikar
Date: April 5, 2018
Time: 11 a.m.
Room: 256C Avery

Abstract: The future pervasive communication and computing devices are envisioned to be tightly integrated with biological systems, i.e., the Internet of Bio-Nano Things. In particular, the study and exploitation of existing processes for the biochemical information exchange and elaboration in biological systems are currently at the forefront of this research direction. Molecular Communication (MC), which studies biochemical information systems with theory and tools from computer communication engineering, has been recently proposed to model and characterize the aforementioned processes. Combined with the rapidly growing field of bio-informatics, which creates a rich profusion of biological data and tools to mine the underlying information, this investigation direction is set to produce interesting results and methodologies not only for systems engineering but also for novel scientific discovery. The multidisciplinary nature of this work presents an interesting challenge in terms of creating a structured approach to combine the aforementioned disciplines for the study of metabolic processes in biological organisms, and their relationship with information for their control, optimization, and exploitation. In this thesis, we study these processes at varying levels of complexity, namely, at the system layer, cellular layer and pathway layer. First, we model the overall functionality of a multicellular metabolic system, the human digestion, in term of energy production from major nutrients in the food. Second, we analyze metabolic processes in a single cell and their adaptability to incoming nutrient availability information form the environment. Third, we model and characterize the processes that enable information to propagate from the external environment and be processed by the cell. Numerical results are presented to provide a first proof-of-concept characterization of all these processes in terms of information theory.




“Consensus Ensemble Approaches Improve De Novo Transcriptome Assemblies”

Adam Voshall

Advisor: Jitender Deogun
Committee members: Juan Cui, Etsuko Moriyama
Date: April 9, 2018
Time: 3:30 p.m.
Room: 256C Avery Hall

Abstract: Accurate and comprehensive transcriptome assemblies lay the foundation for a range of analyses, such as differential gene expression analysis, metabolic pathway reconstruction, novel gene discovery, or metabolic flux analysis. With the arrival of next-generation sequencing technologies it has become possible to acquire the whole transcriptome data rapidly even from non-model organisms. However, the problem of accurately assembling the transcriptome for any given sample remains extremely challenging, especially in species with a high prevalence of recent gene or genome duplications, those with alternative splicing of transcripts, or those whose genomes are not well studied. This thesis provides a detailed overview of the strategies used for transcriptome assembly, including a review of the different statistics available for measuring the quality of transcriptome assemblies with the emphasis on the types of errors each statistic does and does not detect and simulation protocols to computationally generate RNAseq data that present biologically realistic problems such as gene expression bias and alternative splicing. Using such simulated RNAseq data, a comparison of the accuracy, strengths, and weaknesses of seven representative assemblers including de novo, genome-guided methods shows that all of the assemblers individually struggle to accurately reconstruct the expressed transcriptome, especially for alternative splice forms.  Using a consensus of several de novo assemblers can overcome many of the weaknesses of individual assemblers, generating an ensemble assembly with higher accuracy than any individual assembler.




“PhenomeD² - An Integrated Framework For Phenomic Data Analysis”

Venkata Satya Siddhartha Gandha

Advisor: Ashok Samal
Committee members: Vinod Variyam, Jitender Deogun
Date: April 9, 2018
Time: 4:30 p.m.
Room: 347 Avery Hall

Abstract: Knowledge discovery from plant phenotype information along with the genotype data is important for plant scientists, researchers and practitioners. It can lead to better understanding of physiological processes in plants at different stages of growth, under different treatments of nutrients and stress. In the long run, it can lead to development of crops that are tailored to different environments in order to maximize the yield. With the advent of high-throughput phenotyping systems, many experiments with diverse plant species under different growth conditions are being conducted to study phenotypes. These plants are imaged at multiple time points from multiple viewpoints and in different modalities. Researchers are computing a variety of phenotypes (holistic, component; static, dynamic; 2D, 3D) from these images. Currently, there is no mechanism to discover interesting patterns from diverse experiments since there is no unified framework to store the information from them. This project is motivated by the need to organize such high dimensional phenomic data and explore various strategies to analyze, visualize and mining the data.

In this project, we have proposed an integrated framework called Data model and Data analytics on Phenomic data and developed a prototype, PhenomeD². The framework includes a data model that is used to store all the phenomic (and related) information in a relational database. The frontend of PhenomeD² supports: (a) database querying (b) computation of descriptive statistics and visualizations and (c) data mining. These modules provide mechanisms to explore and analyze patterns in the phenomic data. PhenomeD² integrates Weka, a widely used software for data mining. To support visualization of temporal patterns in phenomic data, PhenomeD² uses Dygraphs, a JavaScript-based charting library. The framework supports inclusion of genome sequences and integrating tools to enable QTL/GWAS mappings.




“An Integrated Web Based Platform for Interactive Nitrogen Leaching Exploration”

Babak Samani

Advisor: Hongfeng Yu
Committee members: Haishun Yang, Mohammad Rashedul Hasan, Qiben Yan
Date: April 11, 2018
Time: 10:30 a.m.
Room: 352 Avery Hall

Abstract: While nitrogen (N) is an essential nutrient for corn, its leaching to ground water is an serious environmental issue and a hazard to public health. N leaching is closely linked to soil texture, crop type and weather factors, especially rainfall. Prediction of N leaching in cropping systems is critical to improvement of crop management and reduction of N leaching. The primary contribution of this project is to develop a web based platform that provides an end-to-end solution in support of interactive N leaching estimation and exploration across Nebraska corn fields. At the front end, we design a map based visualization interface that allows a user to interactively specify a region of interest (ROI) and time ranges. Heterogeneous datasets, including weather data, soil textures, and cropland data, within the ROI can be efficiently fetched from our back-end data management system that leverages multiple databases and data store techniques. Maize-N, an N leaching model, takes the fetched data as input and outputs the prediction of N leaching of ROI, which is visualized at the front end. Our design also tackles the security and performance of our system to ensure proper user experience. The platform provides an intuitive and effective tool to facilitate farmers to appropriately manage the usage of N fertilizer and reduce the N leaching in their fields.




“Cost-Effective Techniques for Continuous Integration Testing”

Jingjing Liang

Advisor: Gregg Rothermel
Committee members: Sebastian Elbaum, Witawas Srisa-An
Date: April 12, 2018
Time: 12:30 p.m.
Room: 103C Avery Hall

Abstract: Continuous integration (CI) development environments allow software engineers to frequently integrate and test their code. While CI environments provide advantages, they also utilize non-trivial amounts of time and resources. To address this issue, researchers have adapted techniques for test case prioritization (TCP) and regression test selection (RTS) to CI environments. In general, RTS techniques select test cases that are important to execute, and TCP techniques arrange test cases in orders that allow faults to be detected earlier in testing, providing faster feedback to develop- ers. In this thesis, we provide new TCP and RTS algorithms that make continuous integration processes more cost-effective.

To date, current TCP techniques under CI environments have operated on test suites, and have not achieved substantial improvements. Moreover, they can be in- appropriate to apply when system build costs are high. In this thesis we explore an alternative: prioritization of commits. We use a lightweight approach based on test suite failure and execution history that is highly efficient; our approach “continu- ously” prioritizes commits that are waiting for execution in response to the arrival of each new commit and the completion of each previously commit scheduled for test- ing. We conduct an empirical study on three datasets, and use the APFDC metric to evaluate this technique. The result shows that, after prioritization, our technique can effectively detect failing commits earlier.

To date, current RTS techniques under CI environment is based on two windows in terms of time. But this technique fails to consider the arrival rate of test suites and only takes the results of test suites execution history into account. In this thesis, we present a Count-Based RTS technique, which is based on the test suite failures and execution history by utilizing two window sizes in terms of number of test suites, and a Transition-Based RTS technique, which adds the test suites’ “pass to fault” transitions for selection prediction in addition to the two window sizes. We again conduct an empirical study on three datasets, and use the percentage of faults and percentage of “pass to fault” transition metrics to evaluate these two techniques. The results show that, after selection, Transition-Based technique detects more faults and more “pass to fault” transitions than the existing techniques.