1st Edition

Handbook of Mixed Membership Models and Their Applications

    620 Pages 143 Color Illustrations
    by Chapman & Hall

    620 Pages 143 Color Illustrations
    by Chapman & Hall

    In response to scientific needs for more diverse and structured explanations of statistical data, researchers have discovered how to model individual data points as belonging to multiple groups. Handbook of Mixed Membership Models and Their Applications shows you how to use these flexible modeling tools to uncover hidden patterns in modern high-dimensional multivariate data. It explores the use of the models in various application settings, including survey data, population genetics, text analysis, image processing and annotation, and molecular biology.

    Through examples using real data sets, you’ll discover how to characterize complex multivariate data in:

    • Studies involving genetic databases
    • Patterns in the progression of diseases and disabilities
    • Combinations of topics covered by text documents
    • Political ideology or electorate voting patterns
    • Heterogeneous relationships in networks, and much more

    The handbook spans more than 20 years of the editors’ and contributors’ statistical work in the field. Top researchers compare partial and mixed membership models, explain how to interpret mixed membership, delve into factor analysis, and describe nonparametric mixed membership models. They also present extensions of the mixed membership model for text analysis, sequence and rank data, and network data as well as semi-supervised mixed membership models.

    Mixed Membership: Setting the Stage
    Introduction to Mixed Membership Models and Methods Edoardo M. Airoldi, David M. Blei, Elena A. Erosheva, and Stephen E. Fienberg
    A Tale of Two (Types of) Memberships Jonathan Gruhl and Elena A. Erosheva
    Interpreting Mixed Membership April Galyardt
    Partial Membership and Factor Analysis Zoubin Ghahramani, Shakir Mohamed, and Katherine Heller
    Nonparametric Mixed Membership Models Daniel Heinz

    The Grade of Membership Model and Its Extensions
    A Mixed Membership Approach to Political Ideology Justin H. Gross and Daniel Manrique-Vallier
    Estimating Diagnostic Error without a Gold Standard Elena A. Erosheva and Cyrille Joutard
    Interpretability of Mixed Membership Models Burton H. Singer and Marcia C. Castro
    Mixed Membership Trajectory Models Daniel Manrique-Vallier
    Analysis of Development of Dementia through the Extended TGoM Model Fabrizio Lecci

    Topic Models: Mixed Membership Models for Text
    Bayesian Nonnegative Matrix Factorization with Stochastic Variational Inference John Paisley, David M. Blei, and Michael I. Jordan
    Care and Feeding of Topic Models Jordan Boyd-Graber, David Mimno, and David Newman
    Block-LDA: Jointly Modeling Entity-Annotated Text and Entity-Entity Links Ramnath Balasubramanyan and William W. Cohen
    Robust Estimation of Topic Summaries Leveraging Word Frequency and Exclusivity Jonathan M. Bischof and Edoardo M. Airoldi

    Semi-Supervised Mixed Membership Models
    Mixed Membership Classification for Documents with Hierarchically Structured Labels Frank Wood and Adler Perotte
    Discriminative Mixed Membership Models Hanhuai Shan and Arindam Banerjee
    Mixed Membership Matrix Factorization Lester Mackey, David Weiss, and Michael I. Jordan
    Discriminative Training of Mixed Membership Models Jun Zhu and Eric P. Xing

    Special Methodology for Sequence and Rank Data
    Population Stratification with Mixed Membership Models Suyash Shringarpure and Eric P. Xing
    Mixed Membership Models for Time Series Emily B. Fox and Michael I. Jordan
    Mixed Membership Models for Rank Data Isobel Claire Gormley and Thomas Brendan Murphy

    Mixed Membership Models for Networks
    Hierarchical Mixed Membership Stochastic Blockmodels Tracy M. Sweet, Andrew C. Thomas, and Brian W. Junker
    Analyzing Time-Evolving Networks Qirong Ho and Eric P. Xing
    Mixed Membership Blockmodels for Dynamic Networks with Feedback Yoon-Sik Cho, Greg Ver Steeg, and Aram Galstyan
    Overlapping Clustering Methods for Networks Pierre Latouche, Etienne Birmelé, and Christophe Ambroise

    Subject Index
    Author Index

    References appear at the end of each chapter.


    Edoardo M. Airoldi is an associate professor of statistics at Harvard University. Dr. Airoldi’s current research focuses on statistical theory and methods for designing and analyzing experiments in the presence of network interference as well as on modeling and inferential issues when dealing with network data.

    David M. Blei is a professor of statistics and computer science at Columbia University. Dr. Blei’s research is in statistical machine learning involving probabilistic topic models, Bayesian nonparametric methods, and approximate posterior inference.

    Elena A. Erosheva is an associate professor of statistics and social work at the University of Washington, where she is a core member of the Center for Statistics and the Social Sciences. Dr. Erosheva’s research focuses on the development and application of modern statistical methods to address important issues in the social, medical, and health sciences.

    Stephen E. Fienberg is the Maurice Falk University Professor of Statistics and Social Science at Carnegie Mellon University, where he is co-director of the Living Analytics Research Centre and a member of the Department of Statistics, the Machine Learning Department, the Heinz College, and Cylab. Dr. Fienberg’s research includes the development of statistical methods for categorical data analysis and network data analysis.

     "The editors of this volume have notably worked at the forefront of research in various subfields of mixed membershi modeling since the field’s inception.  . . . The strength of their collaboration in editing this volume can be seen in the book’s organization. . . .One of the main strengths of the book, fulfilling the promise of its title, is the wealth of applications described therein . . . We believe this book sets the stage for a rich body of future work."

    ~Trevor Campbell and Tamara Broderick, Massachusetts Institute of Technology