Neural nonnegative matrix factorization for hierarchical multilayer topic modeling
We introduce a new method based on nonnegative matrix factorization, Neural NMF, for detecting latent hierarchical structure in data. Datasets with hierarchical structure arise in a wide variety of fields, such as document classification, image processing, and bioinformatics. Neural NMF recursively...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer International Publishing
2024
|
| Online Access: | https://hdl.handle.net/1721.1/153309 |
| _version_ | 1826213609332539392 |
|---|---|
| author | Haddock, Jamie Will, Tyler Vendrow, Joshua Zhang, Runyu Molitor, Denali Needell, Deanna Gao, Mengdi Sadovnik, Eli |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Haddock, Jamie Will, Tyler Vendrow, Joshua Zhang, Runyu Molitor, Denali Needell, Deanna Gao, Mengdi Sadovnik, Eli |
| author_sort | Haddock, Jamie |
| collection | MIT |
| description | We introduce a new method based on nonnegative matrix factorization, Neural NMF, for detecting latent hierarchical structure in data. Datasets with hierarchical structure arise in a wide variety of fields, such as document classification, image processing, and bioinformatics. Neural NMF recursively applies NMF in layers to discover overarching topics encompassing the lower-level features. We derive a backpropagation optimization scheme that allows us to frame hierarchical NMF as a neural network. We test Neural NMF on a synthetic hierarchical dataset, the 20 Newsgroups dataset, and the MyLymeData symptoms dataset. Numerical results demonstrate that Neural NMF outperforms other hierarchical NMF methods on these data sets and offers better learned hierarchical structure and interpretability of topics. |
| first_indexed | 2024-09-23T15:51:58Z |
| format | Article |
| id | mit-1721.1/153309 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T15:51:58Z |
| publishDate | 2024 |
| publisher | Springer International Publishing |
| record_format | dspace |
| spelling | mit-1721.1/1533092024-06-28T15:02:17Z Neural nonnegative matrix factorization for hierarchical multilayer topic modeling Haddock, Jamie Will, Tyler Vendrow, Joshua Zhang, Runyu Molitor, Denali Needell, Deanna Gao, Mengdi Sadovnik, Eli Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science We introduce a new method based on nonnegative matrix factorization, Neural NMF, for detecting latent hierarchical structure in data. Datasets with hierarchical structure arise in a wide variety of fields, such as document classification, image processing, and bioinformatics. Neural NMF recursively applies NMF in layers to discover overarching topics encompassing the lower-level features. We derive a backpropagation optimization scheme that allows us to frame hierarchical NMF as a neural network. We test Neural NMF on a synthetic hierarchical dataset, the 20 Newsgroups dataset, and the MyLymeData symptoms dataset. Numerical results demonstrate that Neural NMF outperforms other hierarchical NMF methods on these data sets and offers better learned hierarchical structure and interpretability of topics. 2024-01-11T16:07:30Z 2024-01-11T16:07:30Z 2023-12-19 2023-12-24T04:17:57Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/153309 Sampling Theory, Signal Processing, and Data Analysis. 2023 Dec 19;22(1):4 PUBLISHER_CC en https://doi.org/10.1007/s43670-023-00077-3 Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer International Publishing Springer International Publishing |
| spellingShingle | Haddock, Jamie Will, Tyler Vendrow, Joshua Zhang, Runyu Molitor, Denali Needell, Deanna Gao, Mengdi Sadovnik, Eli Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title | Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title_full | Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title_fullStr | Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title_full_unstemmed | Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title_short | Neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| title_sort | neural nonnegative matrix factorization for hierarchical multilayer topic modeling |
| url | https://hdl.handle.net/1721.1/153309 |
| work_keys_str_mv | AT haddockjamie neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT willtyler neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT vendrowjoshua neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT zhangrunyu neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT molitordenali neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT needelldeanna neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT gaomengdi neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling AT sadovnikeli neuralnonnegativematrixfactorizationforhierarchicalmultilayertopicmodeling |