Hi, my name is Ian Wood. I'm a PhD student in Informatics on the Complex Systems track at Indiana University in Bloomington, IN. I received my B.S. in Computer Science from Clemson University in 2011. I'm interested in complex systems whether theoretical, social, or biological; network science; machine learning; and data analysis in general. This website is a place for me to display some of the dissertation projects I've been working on, and will be changing frequently.
Artificial Immune System Binary Classifier
This is an ongoing project to develop a binary classifier based on a model of T-Cell cross regulation in the immune system.
Personalized Community Detection
This is a work in progress to scale continuous-time markov models for Infomap community detection on networks to large numbers of nodes. By adding exogenous information in the form of rates on the nodes, detected communities can be personalized to the user's task, such as finding groups of similar papers in information retrieval tasks.
Eigenmode Analysis of Emotional Sentiment on Twitter
This work investigates the possibilities of decomposing the distribution of quantified sentiment (using the Affective Norms for English Words) of large numbers of tweets into informative components. Correlating these components with representative days and other collective behavior. Papers should be coming soon.
Towards a Social SLAM
Inspired by the idea of simultaneous localization and mapping in physical space, this project aims to detect the social space (situations) that a household robot with low resolution sensors must operate in. We present a minimalist social robot that relies on long timeseries of low resolution data such as temperature, lighting, sounds and collisions. Our goal is to develop an experimental system for growing socially situated robotic agents whose behavioral repertoire is subsumed by the social order of the space. To get there we are designing robots that use simple sensors and motion feedback routines to recognize different classes of human activity and respond.
We describe exploratory tests that allow us to develop hypotheses about what objects (sensor data) correspond to something known and observable by a human subject.
We use machine learning methods to classify three social scenarios, demonstrating that it is possible to detect social situations with high accuracy, using the low-resolution sensors from our minimalist robot.
Lambda-Calculus Based Model of Autopoiesis
This was a project for my qualifying examination, reproducing Fontana's minimal theory of biology based on randomly-generated, self producing lambda calculus expressions.
In the autopoietic definition of life there are many concepts that are informally defined, resulting in ambiguities that formal systems can help elucidate. One of the most fundamental ambiguities is how processes and components must relate to one another in an autopoietic system. In this paper we discuss the concepts involved in autopoiesis and organizational closure as defined by Maturana and Varela, describe some minimal models that clarify some of these concepts, and finally focus on a lambda calculus based model that makes explicit its assumptions behind the duality between component and process.
Iterated Spatial Prisoner's Dilemma
This class project was a simple genetic algorithm to evolve strategies to the iterated prisoner's dilemma on networks in order to investigate how network topology changes strategies.
This class project simulated a game with rules that lead either to a stable attractor or chaos in order to investigate the entropy of the system.
Undergraduate Honors Thesis: Virtual Tours in Google Street View
In this project we combined a virtual reality headset, a Wiimote, and a custom-built treadmill to create virtual tours in Google Street View.