Mention Recommendation for Twitter with End-to-end Memory Network

Authors: Haoran Huang, Qi Zhang, Xuanjing Huang

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results on a dataset we collected from Twitter demonstrated that the proposed method could outperform stateof-the-art approaches.
Researcher Affiliation Academia Haoran Huang School of Computer Science Fudan University Shanghai, P.R.China huanghr15@fudan.edu.cn; Qi Zhang School of Computer Science Fudan University Shanghai, P.R.China qz@fudan.edu.cn; Xuanjing Huang School of Computer Science Fudan University Shanghai, P.R.China xjhuang@fudan.edu.cn
Pseudocode No The paper describes the model architecture and mathematical formulations but does not provide any structured pseudocode or an algorithm block.
Open Source Code No The paper does not provide any explicit statement or link indicating that the source code for the described methodology is openly available.
Open Datasets No The paper states that the dataset was "constructed from Twitter" and collected by the authors, but it does not provide any concrete access information (link, DOI, specific repository, or citation for public access) for this dataset.
Dataset Splits No The paper states, "Finally, we split the dataset into training and testing sets with an 80/20 ratio." It explicitly mentions training and testing splits, but there is no explicit mention or details provided for a separate validation dataset split.
Hardware Specification Yes The average time cost of recommendation for our model with 6 hops and 300 dimensions is approximately 0.031 seconds, calculated from total cost of 849.511 seconds for 26,653 test instances, which demonstrates that our proposed model is efficient. This was measured on a server with an Nvidia TITAN X graphic card.
Software Dependencies No The paper mentions software for evaluation but does not provide specific version numbers for any software dependencies, libraries, or frameworks used in the experiments.
Experiment Setup Yes The embedding dimension in the experiment was set to 300, and the number of hops was set to 6. The learning rate was set to 0.01, and the dropout rate was set to 0.2.