Semisupervised Autoencoder for Sentiment Analysis
Authors: Shuangfei Zhai, Zhongfei Zhang
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the effectiveness of our model on six sentiment analysis datasets, and show that our model significantly outperforms all the competing methods with respect to classification accuracy. We also show that our model is able to take advantage of unlabeled dataset and get improved performance. |
| Researcher Affiliation | Academia | Shuangfei Zhai, Zhongfei (Mark) Zhang Computer Science Department, Binghamton University 4400 Vestal Pkwy E, Binghamton, NY 13902 szhai2@binghamton.edu zhongfei@cs.binghamton.edu |
| Pseudocode | No | The paper includes mathematical equations and descriptions of methods but does not provide any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code for the described methodology, nor does it include any links to a code repository. |
| Open Datasets | Yes | We evaluate our model on six Sentiment Analysis benchmarks. The first one is the IMDB dataset 1 (Maas et al. 2011)...1http://ai.stanford.edu/ amaas/data/sentiment/. The rest five datasets are all collected from Amazon 2(Blitzer, Dredze, and Pereira 2007)...2http://www.cs.jhu.edu/ mdredze/datasets/sentiment/ |
| Dataset Splits | Yes | Table 1: Statistics of the datasets. IMDB # train 25,000 # test 25,000 # unlabeled 50,000. books # train 10,000 # test 3,105. |
| Hardware Specification | No | The paper does not specify any particular hardware used for running the experiments, such as specific GPU or CPU models, or cloud computing resources. |
| Software Dependencies | No | The paper mentions types of software components like 'SVM2', 'Logistic Regression', 'Re Lu', 'Sigmoid', and 'Stochastic Gradient Descent with momentum', but it does not specify any version numbers for these or other software libraries/frameworks. |
| Experiment Setup | Yes | We use Re Lu max(0, x) as the activation function, and Sigmoid as the decoding function. For SBDAE and NN, a small hidden size is sufficient, so we use 200. For DAE, we observe that it benefits from very large hidden sizes; however, due to computational constraints, we take 2000. All the models are trained with mini-batch Stochastic Gradient Descent with momentum of 0.9. We cross validate β from the set {104, 105, 106, 107, 108}. |