Computational Social Science Institute

University of Massachusetts Amherst
research word cloud
LEONTINE ALKEMA, Assistant Professor, Biostatistics
My primary research focus is the development of statistical models to assess and interpret demographic and population-level health trends and differentials, generally on a national level for all countries globally. I have developed estimation and projection methods for key global indicators, including child and maternal mortality rates, the total fertility rate, and the unmet need for contraceptive methods. One of the main methodological challenges in this area of research arises from the need to capture data-driven trends in data-rich populations, while also producing reliable estimates and projections for populations where data availability is limited and where only proxy indicator data are available. My research has addressed these challenges for various indicators via the conceptualization, development, and validation of context-specific Bayesian models, which usually include hierarchical submodels to allow for "sharing" of information across populations to inform the estimates and projections for data-sparse populations, and the parametrization of biases and measurement error variances to account for data quality issues. I collaborate with various United Nations agencies to make available improved estimation methods and resulting estimates to diverse international audiences.
JAMES ALLAN, Professor, Computer Science
My research spans topics in information retrieval, the technology that underlies search engines and other systems that provide search and organization of large volumes of text. I am generally interested in "information organization" to support a person's or a group’s making sense of and managing information encountered on a daily basis. I am currently exploring those issues using a collection of a million out-of-copyright books, developing methods and tools for supporting browsing and searching within that data. I am also very interested in techniques that support people in critical evaluation of the material they find on the web, and help them understand why a page is educative or why it is not. This set of work is driven by a vision of a tool that shows a person whether the web page in front of them discusses a provocative topic, whether the material is presented in a heavily biased way, whether it represents an outlier (fringe) idea, and how its discussion of issues relates to the broader context and to information presented in authoritative sources.
MICHAEL ASH, Professor, Economics and Public Policy
The Corporate Toxics Information Project analyzes and disseminates information from the US Environmental Protection Agency on corporate releases of toxic chemicals and the resulting exposures of communities to air and water pollution hazards. The Project aims to help community-based activists and socially responsible investors to translate the right to know into the right to clean air and water. The computational methods are wide-ranging, including: fate-and-transport modeling of point-source pollution from industrial facilities; analysis of spatial and socioeconomic data from the Bureau of the Census; and research on corporate structure. Professors Michael Ash and James K. Boyce direct CTIP at the Political Economy Research Institute, UMass Amherst.
PAUL ATTEWELL, Visiting Professor, Sociology.
My research focuses on higher education and particularly issues regarding degree completion among low income students at non-elite colleges. This involves quantitative modeling of longitudinal data from student transcripts. This in turn led to an interest in data mining methods, and a multi-year grant from the National Science Foundation to build an interdisciplinary community and capacity in the use of data mining and Big Data techniques among the social and behavioral sciences and in education.Part of that grant supports graduate students doing research on or using data mining methods; other activities include a seminar series on advances in data mining and analytics, and a graduate-level course on data mining for the social sciences.
SOO YOUNG BAE, Assistant Professor, Communication
I am interested in the far-reaching social and political impact of new communication technologies, with a particular focus on the dynamics of user interaction and information flow in social media platforms. My program of research explores how the flow of news and information becomes increasingly ingrained in existing social relationships, and how it shapes our attitudes, behaviors, and relation to one another.
LAURA BALZAR, Assistant Professor, Biostatistics
I am a methodologist with substantive interests in global health, community-based participatory research, and social determinants of health. My particular areas of expertise are Causal Inference and Machine Learning. My research has focused on challenges that arise when making causal inferences with clustered and longitudinal data, which are often subject to complex measurement, missingness, and dependence. I am the primary statistician for two cluster randomized trials: the SEARCH study for HIV prevention in East Africa and the SPIRIT study for TB prevention in Uganda. Overall, my research is informed by cross-disciplinary real-world problems and aims to ensure methodological advances in Academia translate into real-world impact.
DEEPANKAR BASU, Associate Professor, Economics
My research on economic development looks at issues of undernutrition, health status & health systems, and the condition of informal employment & enterprises in India. I work with data from large scale, nationally representative consumer expenditure surveys conducted by the National Sample Survey Organization (NSSO) to study various aspects of undernutrition and health in India. These include studying the calorie consumption puzzle (as average incomes have risen, average calorie intakes have fallen); the public distribution system; changes in consumption patterns across regions, states, rural & urban areas, and across castes, classes and religions; and the impact of income inequality and public provisioning of health care on average health status across states. I have also been working with firm-level data from the Survey of Unorganized Manufacturing Enterprises conducted by the NSSO to understand the determinants and impacts of subcontracting relationships in the informal sector in India.
RAJESH BHATT, Professor, Linguistics
My research is focused on syntax and the interface between syntax and semantics. I am also interested in computational and mathematical linguistics, in particular in the creation of linguistic resources for computational linguistics such as treebanks. I am an expert on the analysis of the languages of South Asia. I have worked extensively on Hindi-Urdu, Kashmiri, and Kutchi and am interested in the various phenomena that characterize the linguistic situation of the Indian subcontinent. Within syntax and the syntax-semantics interface, I have been interested in relativization, comparatives, modality, and case and agreement. A major long term goal of my research is to harness the linguistic richness of the Indian subcontinent towards a deeper picture of crosslinguistic variation, bringing data from underdocumented and understudied South Asian languages to bear upon our understanding of the human language faculty.
SCOTT BLINDER, Assistant Professor, Political Science
My research focuses on political psychology and political communication around issues involving social identities. This means examining public opinion, media coverage, and political rhetoric--and the relationships among these--in substantive domains such as immigration, integration of religious and ethnic minorities, and gender and politics. Methodologically, I am interested in combining traditional tools of political psychological research with new forms of computer-assisted content analysis of textual data. Substantively, one strand of my research combines public opinion with corpus linguistic research on media coverage of migration and related issues, in an effort to understand the sources of citizens’ implicit beliefs about immigrants and refugees. Another line of research looks at the impact of social norms against prejudice--and individual motivation to follow these norms. I find that these norms and motivations shape shapes attitudes toward diversity and related policies, and also limit responsiveness to attempts to mobilize hostility toward outsiders in service of radical right-wing political movements.
JOSHUA BRAUN, Assistant Professor, Journalism
I am interested, from a sociological perspective, in how infrastructures for media distribution are constructed and maintained. Using conceptual tools from the history and sociology of socio-technical systems, along with lenses from media studies, media sociology, journalism studies, and communication theory, I look at how companies and individuals are building and deploying new systems for distributing their content online. From a methods perspective, I typically use scripting in the service of qualitative research to acquire and sift through data sets in pursuit of interesting cases. I also interview software developers and system administrators about their work on the infrastructures that support media organizations.
DAVID CHIN, Assistant Professor, Health Promotion and Policy
As a health services researcher, I am focused on three domains: 1) the development of outcome measures to quantify quality, value, and safety in healthcare; 2) novel statistical and computational approaches for inference from high-dimensional correlated data; 3) health policy innovation. Unified by data-driven quantitative inference, I emphasize methodology while employing data sets from diverse sources (e.g., electronic health records (EHR), clinical registries, statewide all-payer administrative claims, nationwide private payer, department of defense clinical records) to measure patient outcomes. Some of my ongoing projects include: 1) development of novel instruments to measure Serious Reportable Events occurring in hospitals; 2) the impact of public reporting on cardiovascular procedure outcomes; 3) quantum computing applications for health outcomes and policy.
PAUL M. COLLINS, JR., Associate Professor and Director of Legal Studies, Political Science
The main substantive areas of my research are two-fold. First, I have a keen interest in understanding the democratic nature of the judiciary, particularly relating to interest group litigation, the public’s role in influencing constitutional change, and how presidents attempt to influence and respond to Supreme Court decisions in their public speeches. Second, I study the application of interdisciplinary theories and methods to better understand the behavior of legal actors. To explore these topics, I employ a variety of methodological approaches, including computer assisted content analysis, quantitative data analysis, and historical methods.
BRUCE CROFT, Distinguished Professor, Computer Science
My central research interest is information retrieval – studying representations of information needs and documents in the context of models of relevance. This is the basis of search engines, which are an essential tool for all researchers. One focus of my group is social media such as forums and blogs. We are studying how to search large collections of this type of data, understand the role of conversations, and characterize the topics of conversations on a given topic. Opinion and sentiment analysis is an important topic for many applications related to CSS, and we are working on various aspects of this problem, such as quantifying the diversity and importance of opinions in a ranked list of social posts.
BRIAN DILLON, Assistant Professor, Linguistics
My research is focused on psycholinguistics, the study of how comprehenders understand and produce language in real time. More specifically, I'm interested in how syntactic structures are represented and manipulated in working memory during the course of language comprehension and production. My research involves building explicit computational models of how comprehenders retrieve information from linguistic structures and testing the predictions of those models against actual human linguistic performance. This modeling involves combining features of formal probabilistic parsing models with explicit models of human associative memory to derive predictions about processing speed and accuracy. Specific topics I have investigated involve the computation of agreement in online processing, and the retrieval of referents for pronominal elements. Ultimately, I hope to be able to shed some light on the question of why natural language grammars look the way they do.
ARINDRAJIT DUBE, Associate Professor, Economics
My work focuses on labor economics, health economics, public finance, and political economy. My core areas of research include minimum wage policies, fiscal policy, income inequality, health reform, and the economics of conflict.
LIZ EVANS, Assistant Professor, Health Promotion and Policy
I research how health care systems and public policies can better promote health and wellness among vulnerable and underserved populations, particularly for individuals at risk for opioid and other substance use disorders, mental illness, and infectious diseases. Much of my research has originated from longitudinal study designs and created knowledge via mining of linked administrative data provided by health care delivery systems, social services systems, and criminal justice sources. Most recently, I have investigated the health services utilization and long-term outcomes of patients treated for opioid use disorders; gender differences in the health effects of childhood adversity over the life course; comparative effectiveness of gender sensitive behavioral health care for pregnant and parenting women; and the feasibility of mobile health interventions to reduce infectious disease in global settings. Currently, I am Principal Investigator on a mixed methods study of gender differences among military veterans with chronic musculoskeletal pain in which I am mining electronic health records and utilizing natural language processing to examine patient use of complementary and integrative healthcare and related outcomes (opioid misuse, pain, mental health).
JANE FOUNTAIN, Distinguished Professor, Political Science and Public Policy
Broadly, my fit with the group is as an institutional theorist with longstanding interest and research on the ways that government and related institutions design, enact, implement and use new information and communication technologies. I also examine the policies and practices developed by governments to make use of computational methods. Specifically, my research considers organizational capacity, particularly capacity building across jurisdictional boundaries; legal and regulatory considerations; and the social, political and ethical challenges related to implementation. I currently direct the National Center for Digital Government and, as part of that role and my research program, have been in advisory roles to the U.S. and other governments and international organizations regarding the potential and risks of computational social science in policymaking and governance.
DEEPAK GANESAN, Associate Professor, Computer Science
My research is on wireless sensors and sensor data analytics, most recently the intersection of wireless sensing for health, ultra-low power communication, mobile phone-based sensing, and mobile crowdsourcing. Some of our ongoing work involves holistic sensing of behaviors, for example, designing body sensor systems that can measure activity, gestures, physiology, visual and auditory context, and social interaction continuously in real-world settings to understand and predict addictive behavior such as drug use and smoking. Other ongoing work relates to designing complex wireless sensors with cameras and microphones that operate and communicate solely via harvested energy, such that we no longer need to worry about re-charging our wearables. Another interest of mine is in mobile crowdsourcing, where we analyze task markets that have tens of thousands of participants performing small jobs in the physical world — in this context, we try to understand the interaction between incentives, mobility in the physical world, efficiency, retention and data quality, to help us improve the design of these systems.
INA GANGULI, Assistant Professor, Economics
In my research, I use large datasets to study high-skill labor market issues, with a focus on science and innovation, immigration, and gender topics. Much of my research uses bibliometric data to understand the productivity, mobility and collaboration decisions of scientists. I have also collected unique datasets related to scientific collaborations, including running a web-based survey of authors about their prior scientific collaborations and a field experiment among medical school researchers that involved the collection of nanodata on face-to-face interactions using electronic sensors. Another focus of my research is on understanding gender gaps in education and labor market outcomes using large cross-country datasets. Some of this work uses international, firm-level panel datasets to study gender differentials in the legal and financial sectors. I have also analyzed large, multi-country datasets based on Census and household survey data to understand dynamics in female labor force participation and gender wage gaps.
SHERRY GAO, Associate Professor, Resource Economics
My researches use diversified methods, including mathematical modeling, experimental design and a wide range of statistical methods, to look at questions of methodology importance or policy relevance. I am particularly interested in investigating how people make decisions under risks in the areas of labor economics, health economics, and applied econometrics. In a current project, my coauthors and I are applying Bayesian econometric methods to uncover the decision process that people follow in making choices in simple situations that involve risk. Particularly, we are interested in addressing the heterogeneity among people when making such decisions, and we address this issue using mixture modeling as well as Bayesian hierarchical modeling approaches.
SONG GAO, Associate Professor, Civil and Environmental Engineering
My ultimate research goal is to help build an efficient, reliable and sustainable transportation system. One of the major obstacles to such a system is its inherent uncertainty, due to for example, accidents, inclement weather, work zone construction, and fluctuating demand. The rapid development of sensor, information and communication technologies has made real-time traffic information increasingly available for travelers and system operators to improve decision making in uncertain situations. My research thus focuses on understanding travelers' learning and choice behavior in an uncertain, dynamic, connected, information-rich network and incorporating advanced behavioral theories in transportation network optimization models to improve the performance of transportation systems. I examine the problem of travel choice in an uncertain system through a series of theoretical and empirical studies using data from both laboratory experiments and in-vehicle tracking and monitoring devices in real-life urban networks. I design algorithms for the optimal adaptive routing problem with realistic information accessiblities, including delayed, pre-trip, radio and trajectory information, as well as the optimal non-adaptive path problem where the link-wise and time-wise correlations of link travel times are considered. A current project sponsored by ARPA-E aims to design an incentivizing scheme to award travelers for good choices so that the system-wide congestion, energy consumption and emissions are minimized.
KIM GEISSLER, Assistant Professor, Health Promotion and Policy
I am a health policy researcher using health insurance claims datasets and advanced empirical methods to examine factors affecting access to and quality of health care. My current work focuses on physician referrals and how patient care is coordinated among different physicians treating the same patient. To empirically measure these referrals, I use health insurance claims from nearly the entire non-elderly insured population in Massachusetts. I use economic modeling and social network analysis to examine the effects of organizational affiliations – and changes in these affiliations due to hospital mergers and acquisitions – on physician referral patterns. I also analyze the impacts of such patterns on health care cost and quality.
KRISTA GILE, Assistant Professor, Mathematics and Statistics
In recent years, there has been an explosion in the availability of network data, and a corresponding explosion in studies using such data. Statistical methods for network data have also dramatically increased. Most of these methods, however, assume that the network is fully-observed. In practice, this is often not the case; some network members or ties may not be observed, or the observed network may be only a sampled portion of a larger network. I am a statistician, and most of my work has been aimed at at developing methods for partially-observed network data. I have worked on sampling and missing data issues for classes of statistical network models. Much of my recent work has been aimed at improving inference from data collected through respondent-driven sampling (RDS), a particularly widely-used and challenging form of network sampling. RDS is designed to elicit information from hard-to-reach human populations, often used in high-risk populations such as sex worker and drug users. Sampling begins with a small convenience sample of "seeds." Each participant is then given a small number of uniquely identified coupons to pass to other members of the target population, making them eligible for participation. Sampling proceeds in this manner until the desired sample size is reached. This strategy has been effective at recruiting large diverse samples from challenging populations in which other methods have been unsuccessful. Inference from the resulting samples, however, can be quite challenging. My work involves understanding and elucidating these challenges, and developing improved statistical methods to address them.
PHILLIPA GILL, Assistant Professor, Computer Science
My research interests are in the general area of computer networking and network measurement. Specifically, I aim to use insights gained through measurement to improve the security, reliability and performance of networks. This has led me to focus on measuring online information controls such as censorship, surveillance and network neutrality. I'm also interested in understanding the nature of malware threats targeted at civil society organizations and measuring routing security.
DANIELE GIRARDI, Assistant Professor, Economics
My research focuses on empirical macroeconomics and political economy. I’m currently developing a research project that aims to study the influence of political-institutional factors on macroeconomic variables, using appropriate techniques to avoid reverse causality bias. A second project on which I’m working (joint with Professor Antonella Stirati at Roma Tre University and funded by a grant from the Institute for New Economic Thinking) aims to assess the influence of aggregate demand on economic growth and income distribution in the medium/long-run. From a methodological point of view, I’m interested in the use of explicit research designs and natural experiments for the identification of causal relations in macroeconomics.
SETH K. GOLDMAN, Assistant Professor, Communication and Commonwealth Honors College
I teach and carry out research on the effects of mass media and political communication on stereotyping and prejudice, particularly with regard to public opinion about race and sexual orientation. My recent book, The Obama Effect: How the 2008 Campaign Changed White Racial Attitudes (RSF, 2014), combines large-scale survey data and automated text analysis of all U.S. news sources available on the Internet to demonstrate that mediated exposure to Obama and his family during the 2008 presidential campaign helped to reduce white racial prejudice. As part of this project, I helped develop and execute the 2008 National Annenberg Election Survey (NAES), a five-wave nationally representative panel survey executed online, with 95,464 completed interviews between October 2007 and January 2009. My current research tracks changes over time in racial attitudes beyond the 2008 campaign using additional re-interviews of whites and blacks in 2010 and 2012/2013. In addition to survey research, several of my current projects employ population-based survey experiments. For example, one study uses a national survey and an experiment embedded in a separate survey to investigate the effects of conservatives’ attacks on “activist judges” on the underlying drivers of support for gay marriage. Another study uses two experiments embedded in a nationally representative survey of heterosexual and LGB Americans to assess the effects of “victim” portrayals of LGBs in the media on both majority prejudice and minority pride.
CHAITRA GOPALAPPA, Assistant Professor, Industrial Engineering
My research interests are developing mathematical models for disease prediction, prevention, and control. The mathematical methodologies I most commonly use are from areas of stochastic processes and simulation modeling. I am currently working on two mains areas of application. One is analyzing the impact and cost-effectiveness of alternative portfolio combinations of care interventions for HIV-infected persons, which could help develop an implementation strategy to achieve the 2020 National HIV/AIDS Strategy goals for the United States. This work is in collaboration with the Centers for Disease Control and Prevention. I am also involved in developing natural history models for cancer onset and progression at the national-level, to identify impact and costs of screening interventions, applied to low and middle income countries. This work is in collaboration with the World Health Organization and Avenir Health.
JUSTIN H. GROSS, Assistant Professor, Political Science
My main applied research interests are in mass media and political communication, public opinion, and public policy. In my methodological work, I develop tools of measurement for text, networks, and surveys. I have been collaborating with social and computer scientists to analyze the framing of policy issues and measure communicator ideology within large corpora of text. We seek to combine aspects of qualitatively rich content analysis with modern methods from machine learning and natural language processing, making use of the comparative advantages of both human and computational efforts. I am especially interested in the challenge of detecting abstract ideas in text and measuring their diffusion through communication networks.
PETER M. HAAS, Professor, Political Science.
I am interested in understanding what makes governments cooperate with one another on important global issues (I have worked on many environmental topics – global warming, stratospheric ozone, European acid rain, regional marine pollution, etc.). My interests have to do with the interactions of different political actors (governments, international organizations, private sector, scientific networks, and civil society) in various stages of global governance, and the relative effectiveness of these different patterns. My work has largely been qualitative (process tracing, elite interviews, comparative case studies) but I am also interested in applying network analysis to better map networks of activist scientists and to analyze the interplay between issues and actors.
JOHN A. HIRD, Professor, Political Science and Public Policy
I am a public policy scholar interested in the interplay of expertise and public policymaking. A current project with Bruce Desmarais involves the analysis of how science is employed in regulatory policymaking in the US and, subsequently, the EU, using bibliometric and network analysis. We analyze how regulatory policymakers use science and other forms of information to inform decisionmaking, whether agencies utilize the most appropriate science, and examine the networks of publications, scientists, agency officials, interest groups, etc. that inform a broad range of regulatory policymaking. The project will also focus on how policymakers can better access and use science appropriately in reaching public policy decisions.
DAVID E. HUBER, Professor, Psychological and Brain Sciences
My research focuses on human perception and memory from a broad-based, computational perspective. To shed light on these basic cognitive processes, I find converging evidence from behavioral studies and neurophysiological measures in combination with neural network and Bayesian modeling. Ongoing research topics include recognition/recall memory, the benefits of retrieval practice, metamemory, letter/word perception, face perception, semantics, shifts of attention, and social cognition.
GAJA JAROSZ, Associate Professor, Liguistics
My research program is driven by a central question of cognitive science: what kind of system can explain the process by which children acquire the infinitely expressive and complex system of language from limited exposure to linguistic input? An overarching theme of my research is the development of formal systems that are informed by the insights and results of theoretical linguistics and are also capable of successful and realistic modeling of language acquisition. My approach relies on computational modeling, grounded in probability and statistical learning theory, to develop models of learning, to generate and test predictions of learning theories, and to analyze the relationship between the primary linguistic data and acquired knowledge. I work primarily in the domain of phonology.
DAVID JENSEN, Professor, Computer Science
My research focuses on machine learning and causal inference applied to understanding large systems of interacting components. These can be social systems such as online social networks, technological systems such as large-scale data centers, natural systems such as gene regulatory networks, or systems that combine these elements. Existing computational models of systems such as these tend to fall into two broad categories. The first category, simulation models, require manual construction by knowledgeable researchers, with all the potential biases and error this implies. However, simulation models can represent complex interactions among people and institutions and they can forecast the impact of policy changes. The second category, statistical models, can be built automatically from large amounts of data, often improving on the judgments of experts. However, statistical models usually can represent only relatively simple interactions among people and institutions, and most such models cannot forecast the impact of policy changes because they only capture statistical associations, not causal dependence. My work aims to bridge this gap, by developing data analysis methods for learning causal models of complex social systems. My research draws on a diverse array of recent work in machine learning, graphical models, and automated causal discovery. I focus on developing new underlying theory, novel types of highly expressive statistical models, and new algorithms for automated and semi-automated construction of such models. I also apply these models and algorithms to analyzing the operation of real systems such as social media systems, the US securities industry, the worldwide movie industry, and scientific communities.
JASOM M. KAMILAR, Assistant Professor, Anthropology
My research focuses on understanding the behavioral and ecological variation of primates, especially from evolutionary and geographic perspectives. I use a wide variety of quantitative approaches, including phylogenetic comparative methods, GIS and spatial statistics, ecological modeling, and molecular data analysis. Using these methods allows me to address questions in an integrate and interdisciplinary manner. Some of my current projects include: understanding the factors affecting the distribution and coexistence of primate species, examining the genetics and ecological factors driving variation in primate hair density and length, investigating how local environmental characteristics interact with evolutionary history to produce behavioral variation among and within primate species.
JAMES KITTS, Associate Professor, Sociology
I have contributed to two distinct fields of computational social science. The first CSS field uses mathematical and computational models to understand the dynamics of social networks. This work explores our assumptions, identifies scope conditions, derives new hypotheses for empirical study, and ultimately aims to sharpen our intuitions on the dynamics of the social world. My related work has used simulation to investigate collective implications of communication biases in social networks, identified boundary conditions for the evolution of cooperative social norms, and explored the dynamics of polarization, factionalism, and extremism in social groups. The second CSS field collects and analyzes data on social interaction in networks. For example, I have worked with colleagues in Computer Science to develop methods for collecting social interaction data from recorded audio signals using wearable sensors and location-aware devices, and to apply these methods to study the dynamics of interaction in research teams. I worked on an interdisciplinary team to statistically model the local forces resulting in social segregation of friendships among adolescents in 59 high schools. My newest projects are a study of peer influence on health-related behaviors (diet and exercise) among adolescents in four urban middle schools and a study of the network dynamics of patient exchange among 31 Italian hospitals. The community in CSSI at UMass offers a promising arena for conversations between formal socio-behavioral theory and the new riches of socio-behavioral data.
XIANGRONG KONG, Assistant Professor, Biostatistics
As a biostatistician I have primarily worked in the area of HIV and sexually transmitted disease control in Africa and in the area of public health ophthalmology. Inspired by scientific investigations in these arears, my methodological research focuses on analysis of longitudinal and survival data with complicated correlation structures. A particular area of my current research is to develop an analytical framework to evaluate the population impact of the scale-up of HIV prevention programs. In addition to my biostatistical expertise, I received training in social and behavioral sciences in the context of HIV prevention, to learn theories and methods about behavioral approaches to promote health seeking behaviors, and have intense interests to apply and develop statistical methods for better understanding of social norms and social support in HIV high risk populations.
JAMES KUROSE, Distinguished Professor, Computer Science
RAY LA RAJA, Associate Professor, Political Science
I am a political scientist interested in how political parties, interest groups and citizen activists organize to nominate candidates, win elections and influence government policies. With Bruce Desmarais I published a recent paper on how party networks coordinate to support favorite candidates in congressional elections to help them win. Ongoing work explores how these network dynamics in the financing of elections affect the ideological polarization of political parties in the U.S.. I also am pursuing a future project with graduate student Matthew Denny to examine the potential dynamics of providing every citizen with a publicly-financed voucher that can be used to make contributions to candidates. Additional projects on the topic of political participation include the potential impact on turnout of providing a lottery ticket to any citizen who votes.
MICHAEL LAVINE, Professor, Mathematics and Statistics
Michael Lavine’s primary interests are in statistical theory (Theory of Inference, Mixed Models), with applications in Neuroscience, Ecology, and the Environment. He is also active in consulting across various fields.
JENNIFER LUNDQUIST, Professor, Sociology
I am a social demographer with broad interests in race relations, family, and race/gender stratification. One project with application to CSS methods analyzes interracial interaction among online daters using data acquired from a well known online dating website. The data set consists of nine million registered users worldwide with 200 million messages, from November 2003 to October 2010. Spanning beyond the dyadic interactions of messaging behaviors, I am interested in applications of social network analysis to examine the interconnectedness of daters at the group level and would like to explore how other computational models could be applied to the data.
JON MACHTA, Professor, Physics
I work in computational statistical physics. I am interested in the behavior of simple model systems, such as the Ising spin glass, that display complex behavior resulting from competing, frustrated interactions. My work includes large-scale simulations of these models and the development of new Monte Carlo algorithms for carrying out these simulations. I am also collaborating with ecologists to understand the onset of long range synchrony using relationships between coupled map models of ecosystem dynamics and the Ising model in statistical physics. Finally, I am interested in applying ideas from computer science to classify the complexity of models in statistical physics on the basis of the computational complexity of sampling their states.
BENJAMIN MARLIN, Assistant Professor, Computer Science
My research interests lie at the intersection of artificial intelligence, machine learning and statistics. I am particularly interested in hierarchical probabilistic models and approximate inference/learning techniques. My current research is focused on developing probabilistic models and algorithms for learning from incomplete, uncertain and noisy multivariate time series data. My current research explores applications in both health and behavioral science based on modeling and analysis of electronic health records data as well as data collected from mobile on-body physiological sensors. In the past, I have worked on a number of additional applications including collaborative filtering and ranking, unsupervised structure discovery and feature induction, and object recognition and image labeling.
JENNA MARQUARD, Associate Professor, Mechanical and Industrial Engineering
In my research, I model healthcare providers’ (e.g., physicians’ and nurses’) and patients’ cognitive and behavioral interactions with health information technology (IT) as they make decisions about preventing diseases, diagnosing health conditions, managing chronic diseases, and treating acute illnesses. We often use eye-trackers, observations, and interviews to understand where individuals are looking, and what they are doing and thinking as they perform these tasks. These cognitive and behavioral models can objectively guide health IT design and training. My research sits uniquely at the intersection of health informatics, engineering psychology or human factors, and industrial engineering. My research relies heavily on collaborations with physicians, nurses, medical informaticists, and computer scientists.
ANDREW MCCALLUM, Professor, Computer Science
The main goal of my research is to improve our ability to mine useful knowledge from unstructured text. I am especially interested in information extraction of entities and relations from the Web, large-scale entity resolution, reasoning with uncertainty about databases and crowd-sourced human edits to those databases, understanding the connections between people and between organizations, topic models, expert finding, social network analysis, and mining the scientific research literature & community. Toward this end my group develops and employs various methods in statistical machine learning, natural language processing, information retrieval and data mining---especially probabilistic approaches and graphical models. Among other projects we are currently (a) building digital libraries of scientific research papers and studying scientific community emergence, and (b) creating research systems to support open-access publishing and open peer review, and studying the sociology of alternative scientific peer review systems.
DEBI PRASAD MOHAPATRA, Assistant Professor, Resource Economics
My research is in the area of empirical industrial organization. In particular, I study how structural estimation can guide policy decisions in developing and underdeveloped countries. I am also interested in the empirical study of auctions. My current work focuses on various policies like price control policy and compulsory licensing in the context of the Indian pharmaceutical industry. With careful estimation of structural models and taking into account consumer as well as firm incentives, I show that the Government’s current policies are suboptimal for maximizing consumer welfare, and also propose policies to improve welfare. I have used large panel datasets and combined these with datasets from various other sources to estimate complex models using econometric techniques that include partial identification, set estimation and inference.
SCOTT MONROE, Assistant Professor, College of Education
My research focuses on the development of statistical methods for analyzing educational and psychological test and survey data. In general, I am interested in latent variable modeling, and making inferences about unobserved constructs (e.g., mathematics achievement, anxiety). Such methodology is utilized, for example, by large-scale educational assessments, such as the Programme for International Student Assessment (PISA). In particular, my research concerns the development of methods to support inference, such as goodness-of-fit statistics for model evaluation, and standard errors for characterizing estimation uncertainty. Such tools can help applied social science researchers assess how strongly their substantive theories are supported by a statistical model.
TOM MURRAY, Senior Research Fellow, Computer Science
My current research focus is on supporting higher quality dialogue in online interactions. Currently we are focussing on supporting the skills and behaviors needed to deal successfully with situations involving diverse or conflicting ideas, goals, values, or world-views. We call these the “social deliberative skills” that, in the best cases, citizens and professionals use in contexts with multiple stakeholders. We have been experimenting with software features intended to gently support skills such as self-reflection, perspective taking, and meta-dialogue. We are experimenting with tools that support participants in collaboration and dispute resolution, and are also tools that support facilitators or mediators. As part of this work we are also developing computational techniques for recognizing deliberative skill and other indicators of dialogue quality in online communications. In addition to this recent research I have worked for many years in adaptive learning environment and intelligent tutoring systems.
ANNA NAGURNEY, Professor, Operations and Information Management
My research focuses on network systems from transportation and logistical ones, including supply chains, to financial, economic, and social networks and their integration, along with the Internet. I study and model complex behaviors on a spectrum of critical infrastructure networks with a goal towards providing frameworks and tools for understanding their structure, performance, and resilience, when coupled with human interactions and decision-making. I have contributed to the understanding of the Braess paradox in transportation networks and the Internet. I have also, with doctoral students and collaborators, been researching sustainability and quality issues with applications ranging from pharmaceutical and blood supply chains to perishable food products and fast fashion to humanitarian logistics. My team has advanced methodological and computational tools used in game theory, network theory, equilibrium analysis, and dynamical systems. I am presently a Co-PI on a multi-university NSF grant with UMass Amherst as the lead: Network Innovation Through Choice, which is part of the Future Internet Architecture (FIA) program and am also conducting research on risk management and cybersecurity.
BRENDAN O'CONNOR, Assistant Professor, Computer Science
How do you get insight from a giant pile of documents, given you can't read all of it? I develop natural language processing and machine learning methods to analyze social questions in text corpora, such as news or social media. For example, I've analyzed Twitter to understand how new slang spreads between cities, and how textual sentiment corresponds to public opinion polls. Other applications have looked at censorship in Chinese microblogs and extracting events in international relations from the news. Statistical language patterns can give insight into the underlying social variables (text as measurement); or, they can reveal the socially embedded process of language generation. I am interested in a wide variety of linguistic, computational, and statistical methodologies that are necessary to tackle these questions -- Bayesian inference, optimization, probabilistic graphical models, syntactic parsing, sentiment analysis, crowdsourcing, etc. One current project is a tool for interactive text exploration and visualization (a prototype is available at; I'm looking for new collaborations, which will inform how to best develop this and other methods going forward.
MARK C. PACHUCKI, Assistant Professor, Sociology
I am a sociologist who investigates phenomena at the intersection of social determinants of health, social network dynamics, and culture. While it is commonly accepted that culture and social context reciprocally shape our health, an expanded understanding of how the structure and meanings of relationships can affect individuals' health behaviors is aided by perspectives from computational social science. If we can better understand how individuals are connected during and across different stages of their life, we gain insight into changes in health behaviors and health status at the interpersonal and population level. My current projects make use of wearable sensors and probe how peer and family relationships can influence risky health behaviors.
ANTHONY PAIK, Associate Professor, Sociology
As a sociologist focusing on social networks analysis and formal modeling, I am interested in how computational social science techniques can be used to address empirical problems in novel ways. Currently, I am working on an interdisciplinary project with colleagues from computer science, sociology, and public health to study adolescent cyberaggression, which is often hard to define. This spring, my collaborators and I will develop a smartphone application, which will capture electronic communications and answers to text queries in a sample of adolescent smartphone users. We expect to collect hundreds of thousands of messages and will employ content analysis, natural language processing, and machine learning algorithms to extract and classify cyberaggression content in these communications. We will also employ social network methods to examine the location and the diffusion of cyberaggression content in this network of adolescents.
JOE PATER, Professor, Linguistics
I study the knowledge that people have of the sound system of their language, that is, of its phonology, and how that knowledge is acquired. The models of knowledge and learning that I work with have roots in linguistics, cognitive psychology, and machine learning. I am also interested in the range of variation across the phonologies of the world’s languages, and my students and I have begun to explore how agent-based iterated learning models can help us to better understand the shape of language typology.
MJ PETERSON, Professor, Political Science
Developments in computational and information technology have opened up significant possibilities for analysts of politics and for political actors that cannot be ignored. For analysts, mining “big data,” examining massive amounts of text, and applying computer-based network analysis tools permit eliciting information about interactions that has been impossible to acquire before. This opens up the possibility of better understanding the processes by which political actors create and recreate their worlds through collaboration, competition, and conflict with one another. These new analytical techniques bring two strengths to social science research. First, they promote clear specification of analytical routines, one of the positive results of the increasing discussion of methods in my home discipline of political science. Second, they permit treating particular expressions, choices, and actions in the context of simultaneous activity or interaction among other actors. Acknowledging and tracing the influence of behavior in context will yield scholarship more revealing of the political world. For political actors, computers, information technology, and digital media provide possibilities for effective coordination and cooperation but also another set of weapons in the ongoing struggles between centralizers and decentralizers, rulers and ruled, totalizers and pluralists. The political actor uses are more directly relevant to my own research, which continues to focus on the institutionalized processes by which intergovernmental organizations, transgovernmental networks, and transnational advocacy coalitions affect contemporary global governance, the implications of technological change for particular international regimes, and the ethics that should guide scientists and engineers as they pursue their specialized work and provide expertise for policy making, implementation, and review in a globalized world.
NICHOLAS G REICH, Assistant Professor, Biostatistics
Motivated by global health problems, my research unites principles of statistics with the practice of epidemiology. Active areas of research include developing time series and survival models for the spread of infectious disease, developing statistical methods for the analysis of disease surveillance data, and optimizing design and analysis strategies for cluster-randomized studies.
HENRY RENSKI, Associate Professor, Landscape Architecture and Regional Planning
My research focuses on understanding the forces driving regional economic competitiveness and transformation, and building upon this knowledge to improve the effectiveness of state and local economic development policy. My recent work examines regional influences on entrepreneurship; understanding the evolving spatial patterns of development in the U.S.; and the application of matched longitudinal employer-employee data sets to understand the development of human capital in regional industry clusters. I work extensively with large databases of spatially-referenced information.
JESSE RHODES, Associate Professor, Political Science
Much of my research focuses on how national political institutions and public policies evolve, as well as how these developments shape citizens’ attitudes and political behaviors. I have published extensively on education policy, presidential politics, and party politics, and am currently writing a book on the development of the Voting Rights Act since 1965. While I was trained as a historical institutionalist and make extensive use of historical methods and archival materials in my research, I have increasingly employed computational methods - particularly machine learning to analyze large text datasets - in my research. I believe there are exciting opportunities for synergies between historically-motivated research and computational social science methods.
DOUG RICE, Assistant Professor, Political Studies
My substantive research primarily centers on the influence of judicial institutions on public policy. I approach this question in two ways: first, by understanding what courts do in their opinions, and second, by understanding why the decisions and accompanying opinions were designed as they were. In each case, I develop and employ a variety of tools for the computational analysis of text. As examples, in my recent research I utilize unsupervised topic models in order to understand agenda change and dissenting behavior on the U.S. Supreme Court, deploy machine learning tools to identify media coverage of court decisions as well as interactions between the Supreme Court and Congress, and develop sentiment analysis tools to understand the collegiality of courts.
SHANNON ROBERTS, Assistant Professor, Mechanical and Industrial Engineering
I am a Human Factors engineer focused on transportation safety. My research foci are in three areas: (1) developing driving feedback systems, (2) using computational models to predict driver behavior, and (3) leveraging social influence and social network to change driver behavior. The first research focus is aimed at designing and implementing feedback systems to improve driving behavior, with particular attention paid to teenagers and older adults. Within the context of automated vehicles, I am interested in developing feedback systems that inform drivers of the state of their vehicle as well as other vehicles on the road. The second research focus concerns the implementation of computational models, such as agent-based modeling and exponential random graph models, to predict and evaluate the dynamic effects of behavior change within driving systems. The third research focus aims to leverage social influence and social networking techniques to study and positively change driver behavior. This third research area is particularly relevant for teenage drivers given their frequent use of social media and their rather limited driving expertise.
CHRISTIAN ROJAS, Associate Professor, Resource Economics
My research focuses on how imperfect markets function and the instances in which such imperfections can create public policy concerns. I have studied several market factors that contribute to firms’ ability to charge supra-competitive prices or engage in other anticompetitive behaviors; an important element of my analyses is the modeling of how firms engage and consumers perceive/value product differentiation. More recently, I am interested in modeling consumers’ preferences for nutrients and stockpiling behavior with the aim to better understand the trend for the consumption of less healthy foods; the ultimate objective of his work is to provide recommendations for more effective policy interventions that can curb obesity. Most of my work uses detailed scanner-level data from supermarkets across the US covering dozens of consumer packaged goods over several years. I heavily rely on many computational methods to deal with the large datasets and complex estimation techniques that my work requires.
MEREDITH ROLFE, Assistant Professor, Political Science
RONG RONG, Assistant Professor, Resource Economics
My research revolves around three interconnected topics: causality, social networks and communication. Methodologically, I adopt laboratory and field experiments to draw powerful causal inferences. With randomized control trials, I observe how different incentives and institutions influence individual decisions and shape the network topology. The most intriguing application, to me, involves communication. Whom do I connect to in a communication network? Should I choose to tell (or hide) the truth to those who are connected to me? What happens when individual choices like the above are aggregated to form the communication network? Those questions are at the core of my current inquiry.
BRIAN F. SCHAFFNER, Professor, Political Science
Americans are asked to vote for candidates and choose between policies despite the fact that they often have limited information about these political debates. How do citizens make decisions under these conditions? Which pieces of information become salient to voters in different contexts? Do elites capitalize on these conditions of low information to influence citizens' opinions or vote decisions? And what are the consequences of these dynamics for the quality of representation in the United States? Much of my research focuses on (1) an interest in understanding how citizens think and behave in a context where there is little incentive to be highly informed about politics, (2) the extent to which elites take advantage of this low-information decision making to garner support for themselves and their policies and (3) the consequences of this low-information decision making for representation and accountability in the American political system. In addressing these questions, I have developed an expertise in modern survey research methodology, especially with regard to Internet survey research. I am co-PI for the Cooperative Congressional Election Study (CCES), a large-N data infrastructure project involving the collaboration of hundreds of researchers from over 60 academic institutions. Collectively these research teams have fielded national surveys of 35,000 adults in 2006, 37,000 in 2008, and, with partial funding from NSF, 54,000 in 2010. I am also engaged in research using population data from Catalist, a voter file firm that collects political, demographic, and commercial records on every American adult.
CHARLES SCHWEIK, Professor, Environmental Conservation and Public Policy
I am a social scientist working to understand Internet-based collective action and online commons-based peer production. Over the decade, my research has focused largely on the study of open-source software communities, and the socio-technical systems and governance structures that support these systems of co-production. My recent book, Internet Success: A Study of Open Source Software Commons (MIT Press, 2012) analyzed more than 170,000 such projects, in an effort to explain what leads some to ongoing collaborative success and many others to early abandonment. With this grounding in open source collaboration complete, my research agenda is now expanding into other online peer-production settings that, in some way mimic or borrow collaborative principles from open source software, such as: (1) citizen science in environmental protection and management; (2) open access and open educational resources; (3) open science and collaboration around low-cost or open source engineered scientific equipment and mechanisms to collect and validate data they produce; and (4) the broad study of these and other kinds of “Knowledge Commons” systems.
DANIEL SHELDON, Assistant Professor, Computer Science
I use the tools of networks, probability, and optimization to create algorithms that help us understand and make decisions about the world. My current work focuses on ecology. For example, I develop algorithms to learn models of bird migration at continental scale by leveraging several incomplete but complementary sources of data: bird-watcher observations, weather radar measurements, and acoustic monitoring. My group also develops optimization algorithms to help threatened species by improving connectivity in habitat networks. My current work in computational ecology has a broad methodological intersection with computational social science. I have also investigated CSS topics such as the design of provably robust reputation systems and game-theoretic aspects of link placement in networks.
JOHN R. SIRARD, Assistant Professor, Kinesiology
Broadly, I study youth physical activity and obesity prevention. One research focus is to better understand the predictors of weight-related behaviors (i.e., physical activity, screen time, and dietary patterns). Along with my CSSI research partners (James Kitts, Krista Gile, and Mark Pachucki) our goals are to understand peer influence on youth weight-related behaviors and based on that information, develop prediction models to simulate potential intervention effects, and then implement the most promising intervention strategies in real-world settings.
LAUREL SMITH-DOERR, Professor, Sociology
I am a sociologist of science and technology, and have a broad interest in the science of science. This interest has led me to look at how social networks in science and technology fields operate—from large interfirm networks in for-profit sectors like biotechnology to smaller networks of collaboration on interdisciplinary scientific teams. One current NSF funded project (Women in Science Policy-WiSP) collects administrative data from the federal government that includes all federal employee records from 1994-2008, over 16 million records. I have also been involved in research projects and international workshops on gender equity, science policy, ethics, and the organization and governance of science and technology; as such I have a reflexive research interest in studying the development of computational science as a field (including its demographic composition, organization, etc.). I’m currently serving as the Director of the Institute for Social Science Research and in that role have an interest in engaging in conversation across disciplines and in collaborative projects with the interdisciplinary computational social science community.
CEREN SOYLU, Assistant Professor, Economics and Commonwealth Honors College
My research focuses on the theory of collective action and its applications in social movements and environmental conflicts. Drawing upon an empirical review of cases of collective action and using the theoretical frameworks of game theory and social network analysis, I work on developing a model of collective action as a dynamic process and consider the case in which an individual’s decision to act collectively with fellow group members is influenced by both acquisitive and constitutive motives as well as by their endogenously determined beliefs. Social network analysis is useful in identifying the process of belief formation and the dynamics of the collective action process. Considering different distributions of individuals of different types, I aim to analyze not only quantitative but also qualitative impacts of heterogeneity on collective action. I am also interested in using computer simulation to have a better understanding of these dynamic processes.
JOHN STAUDENMAYER, Professor, Mathematics and Statistics
I have research interests in two areas of statistical methodology, one area of application. I also have a general aspiration to be broadly knowledgeable about many areas of applied statistics. My methodological interests are in measurement error and non-parametric regression (smoothing). I have been especially interested in the links between those areas and mixed / random effect / Bayesian regression models. That research (actually) led to long-term applied collaborations with several groups in Kinesiology and public health. That work has been aimed at developing methods to estimate aspects of physical activity and inactivity from accelerometers and other devices.
MATTHIAS STEINRÜCKEN, Assistant Professor, Biostatistics and Epidemiology
In my research I develop computational and statistical methods to analyze genomic data in (human) populations to better understand the evolutionary forces underlying genetic variation. I am interested in the implications of the increasingly widespread availability of genomic data for personalized medicine. Specifically, some things I work on are the inference of past demographic events in humans from genomic data, the impact of population structure and growth on disease related genetic variation, and identifying beneficial genetic material from ancient DNA or time series genetic data.
FLORENCE SULLIVAN, Associate Professor, Education
The Microgenetic Learning Analytics project aims to develop a new computational method for analyzing face-to-face, collaborative learning group discussions. The goal of microgenetic analysis is to characterize conceptual change over short time periods (minutes, hours, days). The challenge of our work is finding effective means for interpreting spoken language that is often fragmented and referential. While, microgenetic analysis is viewed by educational researchers as one of the most robust methods for understanding how human learning occurs, it is a labor intensive method. For this reason, it is conducted with case study research designs involving one or a few students. Our goal is to develop a computational method that will improve possibilities for performing microgenetic analysis over larger data sets, and to expand the scope of educational research questions that may be addressed with such data sets.
Increasingly scholars have access to dynamic data on employees matched to employers, in some cases for entire economies. This is the administrative side of the big data revolution. I have a current project funded by the NSF Building Data Capacity program to form a network of scholars to develop a metadata and source data archive for the rich organizational data collected by the US Equal Employment Opportunity Commission. I am also developing two-mode network models of organizational dynamics based on mobility of people (edges) between jobs (mode 1) and workplaces (mode 2). Conceptualizing labor markets as the network of transitions among employers may help us understand classic questions of, wage setting, innovation, and local economic development. Initial models focus on all employer-employee matches in the Stockholm labor market from 2001 to 2007. A third project employees administrative longitudinal employer-employee data to describe and model workplace income inequality dynamics in Sweden, Germany, France, and Slovenia. In my work the core assumption is that it is the relationships among actors, rather than the actors themselves, that drive organizational outcomes.
HANNA WALLACH, Assistant Professor, Computer Science
My fundamental research goal is to develop new mathematical models and computational tools for understanding and reasoning about the content, structure, and dynamics of complex social processes. I study a wide range of social processes, ranging from communications between scientists or political leaders to the activities of corporate or governmental organizations. To this end, I develop techniques for aggregating and representing large quantities of data from sources with disparate emphases, methods for analyzing text and network data, robust models for reasoning under uncertain information, and efficient inference algorithms. My research contributes to machine learning, Bayesian statistics, and, in collaboration with social scientists, to the nascent field of computational social science.
NOÉ M. WIENER, Lecturer, Economics
I am an economist interested in income inequality, labor market segmentation and migration. Much of my empirical work consists in the measurement and decomposition of income inequality using survey data. Methodologically, I am interested in statistical and information-theoretical approaches to the micro-macro problem in the social sciences. For instance, I am working on characterizing the distribution of wages as a statistical equilibrium arising from the ceaseless interaction between employers and workers. This approach relies on Maximum Entropy and Bayesian methods to infer unobserved behavioral information from the observed wage outcomes.
LEAH WING, Senior Lecturer, Legal Studies Program, Political Science
I am interested in the relationship between technology and dispute development, prevention, and resolution. In particular, my work focuses on a critical examination of the construction and deployment of power and geographies (physical and virtual) in disputing. Some of the projects I have been working on in multi-disciplinary research teams include exploring the impact of anonymity features in online negotiation, how computational tools can support the use of deliberative skills online, and the ways technology can enhance facilitator interventions in multi-party disputes.
REBECCA WOODLAND, Associate Professor, Educational Policy, Research & Administration
I use social network analysis to explore how connections between actors (most often people) support and/or constrain the spread or flow of innovations, access to important resources, and the attainment of essential organizational outcomes. Currently, I am working with colleagues across campus on an NSF-funded research project to examine school district capacity to implement Computer Science for All (EAGER: CSforALL) in the Springfield and Holyoke Public Schools.
WEIAI WAYNE XU, Assistant Professor, Communication
I am a computational communication researcher. I use digital data to model opinion leadership, engagement and social capital during political events. My current research involves building a predictive model to explain returns in word-of-mouth and community engagement from advocacy groups' strategic social media targeting. I have also been building an intuitive Python & R toolkit for qualitative researchers to make sense of large-scale digital data.
KEVIN YOUNG, Assistant Professor, Political Science
My work involves the application of new methods to the study of interest groups and influence in the making of regulatory policy. In particular I am interested in the ways in which recent advances in quantitative data analysis, such as topic modeling and other automated discourse analysis processes can help to make sense of how interest groups behave and the conditions under which they are most influential. While these tools have been popular among computer scientists for some time they are just beginning to be deployed by political scientists and I am interested in both the advantages and limits of these tools. In my own work on the subject of financial regulation I am currently deploying tools from social network analysis to understand the prevalence of interest group coalitions in financial sector lobbying; topic modeling Federal Reserve speeches to understand how central bankers react to political pressure, and using various other tools of automated discourse analysis to examine the extent of conflict among interest groups as they engage in advocacy over regulatory policy.
HONG YU, Professor, Quantitative Health Sciences, UMass Medical School
My primary research focus is the development of new biomedical informatics and computer science methods to analyze big data in the biomedical and health domain. We develop information retrieval, information extraction, natural language processing, question answering and user interfacing approaches. The detailed research is updated in my homepage
KRISTINE YU, Assistant Professor, Linguistics
My research focuses on speech sounds and how they are patterned in natural language. In particular, I’m interested in aspects of prosody, i.e. (1) tone---the phenomenon of pitch differences distinguishing between word meanings, e.g. in Mandarin, ma with a high level pitch means ‘mother’, but ma with a dipping pitch shape means ‘horse’; and (2) intonation---the phenomenon of pitch differences distinguishing between sentential meanings, e.g. He ate already with a rising pitch shape is a question, but He ate already with a falling pitch shape is a declarative----and (3) how language structure is chunked in the grammar based on these and related phenomena. My research involves the application of machine learning classification algorithms for understanding how tonal and intonational elements are encoded and decoded in the acoustic speech signal as well as building computational models for recovering prosodic grammatical structure from the incoming speech stream.
RODRIGO ZAMITH, Assistant Professor, Journalism
My research focuses on the reconfiguration of journalism in a changing media environment and the development of digital research methods. From a theoretical perspective, I am interested in the changing nature of news and newswork amid the adoption of new technologies, an influx of technologists into newsrooms, and greater emphasis on data. From a methodological perspective, I am interested in developing tools and algorithms to facilitate content analyses of growing datasets. Recently, I have devoted time to developing---and exploring the implications of---a hybrid approach to content analysis that seeks to integrate both manual and automated modes of analysis in order to enhance the validity and reliability of large-scale analyses of media content. I have also recently worked on a process for capturing and automatically coding certain aesthetic properties of websites, and analyzing them over time. Given the interdisciplinarity of my field and my work, I welcome collaborations with fellow scholars whose work intersects with mass communication and/or employs content analysis as a central method.