Network Representation Learning with Rich Text Information
Authors: Cheng Yang, Zhiyuan Liu, Deli Zhao, Maosong Sun, Edward Chang
IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our method and various baseline methods by applying them to the task of multi-class classification of vertices. The experimental results show that, our method outperforms other baselines on all three datasets, especially when networks are noisy and training ratio is small. |
| Researcher Affiliation | Collaboration | Cheng Yang1,2, Zhiyuan Liu1,2 , Deli Zhao2, Maosong Sun1, Edward Y. Chang2 1 Department of Computer Science and Technology, State Key Lab on Intelligent Technology and Systems, National Lab for Information Science and Technology, Tsinghua University, Beijing 100084, China 2 HTC Beijing Advanced Technology and Research Center, China |
| Pseudocode | No | The paper describes algorithmic steps in text but does not include a formally structured pseudocode or algorithm block. |
| Open Source Code | Yes | The source code of this paper can be obtained from https://github.com/albertyang33/TADW. |
| Open Datasets | Yes | We evaluate TADW with five baseline methods of representation learning using three publicly available datasets 1. 1http://linqs.cs.umd.edu/projects//projects/lbc/index.html. |
| Dataset Splits | Yes | The training ratio varies from 10% to 50% for linear SVM and 1% to 10% for TSVM. For each training ratio, we randomly select documents as training set and the remaining documents as test set. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU models, CPU types, or memory specifications used for running experiments. |
| Software Dependencies | No | The paper mentions software like Liblinear and SVM-Light for implementations, but it does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | We set parameters as follows, walks per vertex γ = 80 and window size t = 10 which are the same with those in the original paper. We select k = 80 and λ = 0.2 for Cora and Citeseer datasets; k = 100, 200 and λ = 0.2 for Wiki dataset. |