Microblog Sentiment Classification with Contextual Knowledge Regularization
Authors: Fangzhao Wu, Yangqiu Song, Yongfeng Huang
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on benchmark datasets show that our method can consistently and significantly outperform the state-of-the-art methods. |
| Researcher Affiliation | Academia | Tsinghua National Laboratory for Information Science and Technology, Department of Electronic Engineering, Tsinghua University, Beijing, China University of Illinois at Urbana-Champaign, Urbana, IL, USA |
| Pseudocode | Yes | Algorithm 1 Accelerated algorithm for updating wt+1. |
| Open Source Code | No | The paper does not provide an unambiguous statement or link for the open-source code of the described methodology. |
| Open Datasets | Yes | The first dataset is Sanders Twitter sentiment dataset1. This dataset contains 3,727 hand-labeled tweets related to four companies: Apple, Goolge, Twitter and Microsoft. [footnote 1: http://www.sananalytics.com/lab/twitter-sentiment/] The second dataset is the test data in Stanford sentiment corpus2 (denoted as STS-manual) which was labeled manually. [footnote 2: http://help.sentiment140.com/for-students] The third dataset is Twitter sentiment classification dataset provided by Sem Eval 2013 conference3 (denoted as Sem Eval) [footnote 3: http://www.cs.york.ac.uk/semeval-2013/task2/]. In order to extract the contextual knowledge from unlabeled data, we used a large dataset, i.e., the training data in Stanford sentiment corpus4 (denoted as STS-emoticon). [footnote 4: http://help.sentiment140.com/for-students] |
| Dataset Splits | Yes | Five-fold cross-validation was used for STS and Sanders datasets. For Sem Eval dataset, the original splitting was used. Parameters were tuned on the validation sets. |
| Hardware Specification | No | The paper does not explicitly describe the specific hardware (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific software names with version numbers for its ancillary software dependencies. |
| Experiment Setup | No | The paper mentions that "Parameters were tuned on the validation sets" and details preprocessing steps, but it does not provide specific hyperparameter values (e.g., learning rate, batch size, epochs) or detailed system-level training configurations for model training. |