Exploring Multiple Feature Spaces for Novel Entity Discovery

Authors: Zhaohui Wu, Yang Song, C. Giles

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on three real-world datasets show that our method significantly outperforms existing methods on identifying novel entities.
Researcher Affiliation Collaboration Zhaohui Wu , Yang Song , C. Lee Giles Computer Science and Engineering, Information Sciences and Technology Pennsylvania State University, University Park, PA 16802, USA Microsoft Research, One Microsoft Way, Redmond, WA 98052, USA zzw109@psu.edu, yangsong@microsoft.com, giles@ist.psu.edu
Pseudocode No The paper describes its methods in prose but does not include any explicit pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement or a link to the source code for the methodology described.
Open Datasets Yes We use three datasets for our evaluation: 1) AIDA-EE dataset from Hoffart et al. (2014), 2) Web News dataset from a Web news portal, and 3) Wikievents dataset from Wikipedia event news.7 All three datasets use Wikipedia as the knowledge base for evaluation. ... 7http://www.cse.psu.edu/ zzw109/data.html
Dataset Splits No The paper describes training and test splits for the datasets (e.g., "training dataset and ... test dataset" for AIDA-EE, "used for training while ... used for testing" for Web News and Wiki Events), but does not explicitly mention a separate validation dataset split with specific details for their experiments.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory specifications) used for running its experiments.
Software Dependencies Yes We use Stanford Dependencies4 that represent the grammatical relations between words using triplets: name of relation, governor and dependent, generated by the Stanford Parser 3.4 (Socher et al. 2013).
Experiment Setup Yes setting Num Trees=100, Num Leaves=20, Min Instances In Leaf=10, and Learning Rate=0.2.