Acronym Disambiguation Using Word Embedding
Authors: Chao Li, Lei Ji, Jun Yan
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the models on MSH Dataset and Science WISE Dataset, and both models outperform the state-of-art methods on accuracy. The experimental results show that word embedding helps to improve acronym disambiguation. |
| Researcher Affiliation | Collaboration | Chao Li1*, Lei Ji2, Jun Yan2 1 Dalian University of Technology, Dalian, China, 2 Microsoft Research, Beijing, China oahcil.dlut@gmail.com, leiji@microsoft.com, junyan@microsoft.com |
| Pseudocode | No | The paper provides mathematical formulas for TBE and SBE but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement about releasing open-source code or a link to a code repository. |
| Open Datasets | Yes | We evaluate our models on MSH Collection and Science WISE Collection shared by (Prokofyev et al. 2013) which are both collections of scientific abstracts containing ambiguous acronyms. |
| Dataset Splits | No | The MSH dataset contains 7,641 abstracts in training data and 3,631 abstracts in test data, while the Science WISE dataset contains 2,943 abstracts in training data and 2,267 abstracts in test data. No explicit mention of a validation set split was found. |
| Hardware Specification | No | The paper does not provide any specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using word embedding based on (Mikolov et al. 2013) but does not provide specific version numbers for any software dependencies, libraries, or frameworks used for implementation. |
| Experiment Setup | Yes | Then, we apply our models on the datasets to get the embeddings of acronyms (in TBE we use top 20 TF-IDF words, in SBE the word window is 3). |