Enriching Documents with Compact, Representative, Relevant Knowledge Graphs
Authors: Shuxin Li, Zixian Huang, Gong Cheng, Evgeny Kharlamov, Kalpa Gunaratna
IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments and user studies show the promising performance of our approach. |
| Researcher Affiliation | Collaboration | Shuxin Li1 , Zixian Huang1 , Gong Cheng1 , Evgeny Kharlamov2,3 and Kalpa Gunaratna4 1National Key Laboratory for Novel Software Technology, Nanjing University, China 2Bosch Center for Artificial Intelligence, Robert Bosch Gmb H, Germany 3Department of Informatics, University of Oslo, Norway 4Samsung Research America, Mountain View CA, USA |
| Pseudocode | Yes | Algorithm 1 Computation of Qmax |
| Open Source Code | Yes | Source code and data: https://github.com/nju-websoft/CR2. |
| Open Datasets | Yes | Documents. We ran our experiments over documents sampled from the Signal Media One-Million News Articles Dataset (Signal-1M) [Corney et al., 2016]. |
| Dataset Splits | No | The paper mentions the number of documents used for experiments (e.g., '100 documents', '6,400 documents') but does not specify explicit train/validation/test dataset splits with percentages or sample counts. |
| Hardware Specification | Yes | Our experiments were performed on an Intel Xeon E5-1607 (3.10 GHz) with 40GB memory for Java. |
| Software Dependencies | No | The paper mentions 'Java' and 'DBpedia Spotlight' but does not provide specific version numbers for any software dependencies or libraries. |
| Experiment Setup | Yes | Configuration of CR2. We set the diameter bound D = 4 following Cheng et al. [2017]. For Eq. (3), we set α = 0.5 but we may obtain better results by tuning this parameter. For Eq. (10), we set k = 0.4 following Zhu et al. [2017]. |