Probabilistic Models for Scalable Knowledge Graph Construction

In the past decade, systems that extract information from millions of Internet documents have become commonplace. Knowledge graphs -- structured knowledge bases that describe entities, their attributes and the relationships between them -- are a powerful tool for understanding and organizing this vast amount of information. However, a significant obstacle to knowledge graph construction is the unreliability of the extracted information, due to noise and ambiguity in the underlying data or errors made by the extraction system and the complexity of reasoning about the dependencies between these noisy extractions. My dissertation addresses these challenges by exploiting the interdependencies between facts to improve the quality of the knowledge graph in a scalable framework. I introduce a new approach called knowledge graph identification (KGI), which resolves the entities, attributes and relationships in the knowledge graph by incorporating uncertain extractions from multiple sources, entity co-references, and ontological constraints. I define a probability distribution over possible knowledge graphs and infer the most probable knowledge graph using a combination of probabilistic and logical reasoning. Such probabilistic models are frequently dismissed due to scalability concerns, but my implementation of KGI maintains tractable performance on large problems through the use of hinge-loss Markov random fields, which have a convex inference objective. This allows the inference of large knowledge graphs using 4M facts and 20M ground constraints in 2 hours. To further scale the solution, I develop a distributed approach to the KGI problem which runs in parallel across multiple machines, reducing inference time by 90%. Finally, I extend my model to the streaming setting, where a knowledge graph is continuously updated by incorporating newly extracted facts. I devise a general approach for approximately updating inference in convex probabilistic models, and quantify the approximation error by defining and bounding inference regret for online models. Together, my work retains the attractive features of probabilistic models while providing the scalability necessary for large-scale knowledge graph construction. These models have been applied on a number of real-world knowledge graph projects, including the NELL project at Carnegie Mellon and the Google Knowledge Graph.

A Figured Worlds Perspective on Middle School Learners' Climate Literacy Development

The purpose of this study was to investigate the nature of the relationship between middle school science learners’ conditions and their developing understandings of climate change. I applied the anthropological theoretical perspective of figured worlds (Holland, Lachicotte, Skinner, & Cain, 1998) to examine learners’ views of themselves and their capacities to act in relation to climate change. My overarching research question was: How are middle school science learners’ figured worlds of climate change related to the conditions in which they are embedded? I used a descriptive single-case study design to examine the climate change ideas of eight purposefully selected 6th grade science learners. Data sources included: classroom observations, curriculum documents, interviews, focus groups, and written assessments and artifacts, including learners’ self- generated drawings. I identified six analytic lenses with which to explore the data. Insights from the application of these analytic lenses provided information about the elements of participants’ climate change stories, which I reported through the use of a storytelling heuristic. I then synthesized elements of participants’ collective climate change story, which provided an “entrance” (Kitchell, Hannan, & Kempton, 2000, p. 96) into their figured world of climate change. Aspects of learners’ conditions—such as their worlds of school, technology and media use, and family—appeared to shape their figured world of climate change. Within their figured world of climate change, learners saw themselves—individually and as members of groups—as inhabiting a variety of climate change identities, some of which were in conflict with each other. I posited that learners’ enactment of these identities – or the ways in which they expressed their climate change agency – had the potential to reshape or reinforce their conditions. Thus, learners’ figured worlds of climate change might be considered “spaces of authoring” (Holland et al., 1998, p. 45) with potential for inciting social and environmental change. The nature of such change would hinge on the extent to which these nascent climate change identities become salient for these early adolescent learners through their continued climate change learning experiences. Implications for policy, curriculum and instruction, and science education research related to climate change education are presented.

DESIGN, MODELING, AND FABRICATION OF MICROROBOT LEGS

This dissertation presents work done in the design, modeling, and fabrication of magnetically actuated microrobot legs. Novel fabrication processes for manufacturing multi-material compliant mechanisms have been used to fabricate effective legged robots at both the meso and micro scales, where the meso scale refers to the transition between macro and micro scales. This work discusses the development of a novel mesoscale manufacturing process, Laser Cut Elastomer Refill (LaCER), for prototyping millimeter-scale multi-material compliant mechanisms with elastomer hinges. Additionally discussed is an extension of previous work on the development of a microscale manufacturing process for fabricating micrometer-sale multi-material compliant mechanisms with elastomer hinges, with the added contribution of a method for incorporating magnetic materials for mechanism actuation using externally applied fields. As both of the fabrication processes outlined make significant use of highly compliant elastomer hinges, a fast, accurate modeling method for these hinges was desired for mechanism characterization and design. An analytical model was developed for this purpose, making use of the pseudo rigid-body (PRB) model and extending its utility to hinges with significant stretch component, such as those fabricated from elastomer materials. This model includes 3 springs with stiffnesses relating to material stiffness and hinge geometry, with additional correction factors for aspects particular to common multi-material hinge geometry. This model has been verified against a finite element analysis model (FEA), which in turn was matched to experimental data on mesoscale hinges manufactured using LaCER. These modeling methods have additionally been verified against experimental data from microscale hinges manufactured using the Si/elastomer/magnetics MEMS process. The development of several mechanisms is also discussed: including a mesoscale LaCER-fabricated hexapedal millirobot capable of walking at 2.4 body lengths per second; prototyped mesoscale LaCER-fabricated underactuated legs with asymmetrical features for improved performance; 1 centimeter cubed LaCER-fabricated magnetically-actuated hexapods which use the best-performing underactuated leg design to locomote at up to 10.6 body lengths per second; five microfabricated magnetically actuated single-hinge mechanisms; a 14-hinge, 11-link microfabricated gripper mechanism; a microfabricated robot leg mechansim demonstrated clearing a step height of 100 micrometers; and a 4 mm x 4 mm x 5 mm, 25 mg microfabricated magnetically-actuated hexapod, demonstrated walking at up to 2.25 body lengths per second.

Reclaiming the Fall Zone: Mediating Physical and Cultural Exchange in Richmond, VA

This thesis will address cultural and physical place reclamation, at the ambiguous intersection of ‘city’ and nature.’ By creating a juxtaposed sequence of multi-scalar interventions, which challenge the conventional boundaries of architecture, and landscape architecture; in order to make commonplace a new dynamic threshold condition in Richmond, Virginia. At its core, this thesis is an attempt at place-making on a site which has become ‘no place.’ This concept will be manifest via a landscape park on Mayo Island in Richmond, anchored by a community retreat center, and architectural follies along a constructed path. The interventions will coincide with value of place in historical Richmond: an integrated, socially desegregated waterfront hinge; a social nexus of inherent change, at the point which the river itself changes at the fall line.