We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: The Application of Active Query K-Means in Text Classification

Authors: Yukun Jiang
Abstract: Active learning is a state-of-art machine learning approach to deal with an abundance of unlabeled data. In the field of Natural Language Processing, typically it is costly and time-consuming to have all the data annotated. This inefficiency inspires out our application of active learning in text classification. Traditional unsupervised k-means clustering is first modified into a semi-supervised version in this research. Then, a novel attempt is applied to further extend the algorithm into active learning scenario with Penalized Min-Max-selection, so as to make limited queries that yield more stable initial centroids. This method utilizes both the interactive query results from users and the underlying distance representation. After tested on a Chinese news dataset, it shows a consistent increase in accuracy while lowering the cost in training.
Comments: 6 pages, 3 algorithms, 4 tables, 8 figures For source code and questions, please email Yukun Jiang at jy2363@nyu.edu Reply would follow shortly
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG); Statistics Theory (math.ST)
Journal reference: In proceedings of 3rd International Conference on Natural Language Processing 2021
Cite as: arXiv:2107.07682 [cs.CL]
  (or arXiv:2107.07682v1 [cs.CL] for this version)

Submission history

From: Yukun Jiang [view email]
[v1] Fri, 16 Jul 2021 03:06:35 GMT (630kb)

Link back to: arXiv, form interface, contact.