A Topic Modeling Approach to Rank Aggregation

We propose a new model for rank aggregation from pairwise comparisons that captures both ranking heterogeneity across users and ranking inconsistency for each user. We establish a formal statistical equivalence between the new model and topic models. We leverage recent advances in the topic modeling literature to develop an algorithm that can learn shared latent rankings with provable statistical and computational efficiency guarantees. The method is also shown to empirically outperform competing approaches on some semi-synthetic and real world datasets.

[1]  M. Kendall.  A NEW MEASURE OF RANK CORRELATION , 1938 .

[2]  Devavrat Shah,et al.  Inferring rankings under constrained sensing , 2008, NIPS.

[3]  Ruslan Salakhutdinov,et al.  Evaluation methods for topic models , 2009, ICML '09.

[4]  Devavrat Shah,et al.  A Data-Driven Approach to Modeling Choice , 2009, NIPS.

[5]  Tao Qin,et al.  A New Probabilistic Model for Rank Aggregation , 2010, NIPS.

[6]  David B. Dunson,et al.  Probabilistic topic models , 2011, KDD '11 Tutorials.

[7]  Craig Boutilier,et al.  Learning Mallows Models with Pairwise Preferences , 2011, ICML.

[8]  David F. Gleich,et al.  Rank aggregation via nuclear norm minimization , 2011, KDD.

[9]  Ashish Kapoor,et al.  Efficient Probabilistic Inference with Partial Ranking Queries , 2011, UAI.

[10]  Sanjeev Arora,et al.  Learning Topic Models -- Going beyond SVD , 2012, 2012 IEEE 53rd Annual Symposium on Foundations of Computer Science.

[11]  Devavrat Shah,et al.  Iterative ranking from pair-wise comparisons , 2012, NIPS.

[12]  Sanjeev Arora,et al.  A Practical Algorithm for Topic Modeling with Provable Guarantees , 2012, ICML.

[13]  Stanley Osher,et al.  Enhanced statistical rankings via targeted data collection , 2013, ICML.

[14]  Nebojsa Jojic,et al.  Efficient Ranking from Pairwise Comparisons , 2013, ICML.

[15]  David C. Parkes,et al.  Generalized Random Utility Models with Multiple Types , 2013, NIPS.

[16]  Venkatesh Saligrama,et al.  Topic Discovery through Data Dependent and Random Projections , 2013, ICML.

[17]  L. Bottou,et al.  Generalized Method-of-Moments for Rank Aggregation , 2013 .

[18]  Arun Rajkumar,et al.  A Statistical Convergence Perspective of Algorithms for Rank Aggregation from Pairwise Data , 2014, ICML.

[19]  Venkatesh Saligrama,et al.  Efficient Distributed Topic Modeling with Provable Guarantees , 2014, AISTATS.

[20]  Avrim Blum,et al.  Learning Mixtures of Ranking Models , 2014, NIPS.

[21]  Devavrat Shah,et al.  Learning Mixed Multinomial Logit Model from Ordinal Data , 2014, NIPS.

[22]  Maksims Volkovs,et al.  New learning methods for supervised and unsupervised preference aggregation , 2014, J. Mach. Learn. Res..