Summary of Document Set Expansion with Positive-unlabeled Learning: a Density Estimation-based Approach, by Haiyang Zhang et al.
Document Set Expansion with Positive-Unlabeled Learning: A Density Estimation-based Approach
by Haiyang Zhang, Qiuyi Chen, Yuanjie Zou, Yushan Pan, Jia Wang, Mark Stevenson
First submitted to arxiv on: 20 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Information Retrieval (cs.IR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel framework, puDE, for document set expansion (DSE), which identifies relevant documents from a large collection based on a small set of documents on a fine-grained topic. Building upon previous work using positive-unlabelled learning (PU) methods, the authors address unresolved issues such as unknown class prior and imbalanced data, and the need for transductive experimental settings. The puDE framework is density estimation-based and does not require class prior knowledge or specific assumptions like SCAR. The method’s effectiveness is demonstrated on real-world datasets, making it a promising alternative for DSE tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In simple terms, this research helps computers find relevant documents from a huge collection based on a small set of similar documents. It solves problems that previous methods had, such as not knowing how many important documents there are or having an unequal number of important and unimportant documents. The new method is better because it doesn’t need to make assumptions about the data or know which class is most common. This makes it useful for finding relevant documents in many areas. |
Keywords
* Artificial intelligence * Density estimation
