End-to-End Multi-Perspective Matching for Entity Resolution
Authors: Cheng Fu, Xianpei Han, Le Sun, Bo Chen, Wei Zhang, Suhui Wu, Hao Kong
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on two real-world datasets show that our method significantly outperforms previous ER methods. |
| Researcher Affiliation | Collaboration | Cheng Fu1,3*, Xianpei Han1,2*, Le Sun1,2, Bo Chen1, Wei Zhang4, Suhui Wu4 and Hao Kong4 1Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences 2State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences 3University of Chinese Academy of Sciences 4Alibaba Group, China |
| Pseudocode | No | The paper describes the model architecture and components in detail but does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper does not contain any statements about making its source code available or provide a link to a code repository. |
| Open Datasets | Yes | We conduct experiments on two real-world datasets: Walmart-Amazon [Konda et al., 2016] and Amazon-Google [Köpcke et al., 2010]. |
| Dataset Splits | Yes | For model learning, we use the same 60%/20%/20% train/dev/test split as in [Mudgal et al., 2018], and use Adam algorithm [Kingma and Ba, 2014] for optimization. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as GPU or CPU models, or memory specifications. |
| Software Dependencies | No | The paper mentions software components like 'FastText', 'Bi-GRU', and 'Adam algorithm' but does not provide specific version numbers for any of them. |
| Experiment Setup | Yes | For our system, we use the pretrained Fast Text 300-dimensional word embedding [Bojanowski et al., 2016] for the two DL based similarity measures: rnn_sim and hybrid_sim. Hidden size of each GRU layer is set 256. For model learning, we use the same 60%/20%/20% train/dev/test split as in [Mudgal et al., 2018], and use Adam algorithm [Kingma and Ba, 2014] for optimization. We use three model settings in our experiments: MPM-ave, MPM-soft, MPM. |