Semantic Visualization with Neighborhood Graph Regularization
Authors: Tuan M. V. Le, Hady W. Lauw
JAIR 2016 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To validate the efficacy of Semafore, our comprehensive experiments on a number of real-life text datasets of news articles and Web pages show that the proposed methods outperform the state-of-the-art baselines on objective evaluation metrics. |
| Researcher Affiliation | Academia | Tuan M. V. Le EMAIL Hady W. Lauw EMAIL School of Information Systems Singapore Management University 80 Stamford Road, Singapore 178902 |
| Pseudocode | No | The paper describes the EM algorithm steps and gradient computations in text and mathematical equations in Section 5 'Model Fitting', but it does not present them in a structured pseudocode or algorithm block. |
| Open Source Code | No | The paper does not explicitly state that source code is provided or offer any links to a code repository. |
| Open Datasets | Yes | We use three real-life, publicly available datasets (Cardoso-Cachopo, 2007) for evaluation. 20News contains newsgroup articles (in English) from 20 classes. Reuters8 contains newswire articles (in English) from 8 classes. Cade12 contains web pages (in Brazilian Portuguese) classified into 12 classes. |
| Dataset Splits | No | The paper mentions creating balanced classes by sampling documents from each class and describes an evaluation method where a document's true class is 'hidden' for prediction using nearest neighbors. However, it does not specify explicit training, validation, and test dataset splits in the traditional sense for reproducing the data partitioning for model training. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not specify any particular software libraries, frameworks, or their version numbers used in the implementation. |
| Experiment Setup | Yes | We set the hyper-parameters to α = 0.01, β = 0.1N and γ = 0.1Z following PLSV (Iwata et al., 2008). In the E-step, P(z|n, m, ˆΨ) is updated as follows: ... In the M-step, by maximizing Q(Ψ|ˆΨ) w.r.t θzw, the next estimate of word probability θzw is as follows: ... φz and xn cannot be solved in a closed form, and are estimated by maximizing Q(Ψ|ˆΨ) using quasi-Newton (Liu & Nocedal, 1989). |