%0 Journal Article %1 blei2010nested %A Blei, David M. %A Griffiths, Thomas L. %A Jordan, Michael I. %C New York, NY, USA %D 2010 %I ACM %J J. ACM %K chinese_restaurant_process clustering combinatorial_stochastic_processes nonparametric_bayesian %N 2 %P 7:1--7:30 %R 10.1145/1667053.1667056 %T The Nested Chinese Restaurant Process and Bayesian Nonparametric Inference of Topic Hierarchies %U http://doi.acm.org/10.1145/1667053.1667056 %V 57 %X We present the nested Chinese restaurant process (nCRP), a stochastic process that assigns probability distributions to ensembles of infinitely deep, infinitely branching trees. We show how this stochastic process can be used as a prior distribution in a Bayesian nonparametric model of document collections. Specifically, we present an application to information retrieval in which documents are modeled as paths down a random tree, and the preferential attachment dynamics of the nCRP leads to clustering of documents according to sharing of topics at multiple levels of abstraction. Given a corpus of documents, a posterior inference algorithm finds an approximation to a posterior distribution over trees, topics and allocations of words to levels of the tree. We demonstrate this algorithm on collections of scientific abstracts from several journals. This model exemplifies a recent trend in statistical machine learning—the use of Bayesian nonparametric methods to infer distributions on flexible data structures.

@article{blei2010nested, abstract = {We present the nested Chinese restaurant process (nCRP), a stochastic process that assigns probability distributions to ensembles of infinitely deep, infinitely branching trees. We show how this stochastic process can be used as a prior distribution in a Bayesian nonparametric model of document collections. Specifically, we present an application to information retrieval in which documents are modeled as paths down a random tree, and the preferential attachment dynamics of the nCRP leads to clustering of documents according to sharing of topics at multiple levels of abstraction. Given a corpus of documents, a posterior inference algorithm finds an approximation to a posterior distribution over trees, topics and allocations of words to levels of the tree. We demonstrate this algorithm on collections of scientific abstracts from several journals. This model exemplifies a recent trend in statistical machine learning—the use of Bayesian nonparametric methods to infer distributions on flexible data structures.}, acmid = {1667056}, added-at = {2014-03-04T16:07:00.000+0100}, address = {New York, NY, USA}, articleno = {7}, author = {Blei, David M. and Griffiths, Thomas L. and Jordan, Michael I.}, biburl = {https://www.bibsonomy.org/bibtex/2ccbe96f63d46203b3729be3d239861bd/peter.ralph}, doi = {10.1145/1667053.1667056}, interhash = {ba7a90e3225576658069310fac939fe7}, intrahash = {ccbe96f63d46203b3729be3d239861bd}, issn = {0004-5411}, issue_date = {January 2010}, journal = {J. ACM}, keywords = {chinese_restaurant_process clustering combinatorial_stochastic_processes nonparametric_bayesian}, month = feb, number = 2, numpages = {30}, pages = {7:1--7:30}, publisher = {ACM}, timestamp = {2014-03-04T16:07:00.000+0100}, title = {The Nested Chinese Restaurant Process and Bayesian Nonparametric Inference of Topic Hierarchies}, url = {http://doi.acm.org/10.1145/1667053.1667056}, volume = 57, year = 2010 }