From Tweets to Wellness: Wellness Event Detection from Twitter Streams
Authors: Mohammad Akbari, Xia Hu, Nie Liqiang, Tat-Seng Chua
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on a real-world dataset from Twitter have demonstrated the superior performance of our framework. |
| Researcher Affiliation | Academia | Mohammad Akbari a,b, Xia Huc, Nie Liqiangb, Tat-Seng Chua a,b a NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore b School of Computing, National University of Singapore, Singapore c Department of Computer Science & Engineering, Texas A&M University, USA |
| Pseudocode | Yes | Algorithm 1: Optimization algorithm of Eq. (5) |
| Open Source Code | No | We plan to release this dataset to facilitate others in reproducing our experiments as well as verifying their ideas.2 Available at www.comp.nus.edu.sg/ a0103416. (The URL is stated to contain the dataset, not the source code for the methodology.) |
| Open Datasets | Yes | We construct a large-scale diabetes dataset by automatically extracting lifestyle and wellness related short messages and manually constructing the ground-truth labels. We plan to release this dataset to facilitate others in reproducing our experiments as well as verifying their ideas.2 Available at www.comp.nus.edu.sg/ a0103416. |
| Dataset Splits | Yes | We divided the dataset into two sets based on their posting times. In particular, tweets that were posted before May 2015 were utilized to train our model; while those posted from May to July 2015 were used for evaluation process. and For each method mentioned above, the respective parameters were carefully tuned based on 5-fold cross validation on the training set and the parameters with the best performance were used to report the final comparison results. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU, GPU models, or memory specifications) used for running experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python, library versions, or specific solver versions) required to replicate the experiments. |
| Experiment Setup | Yes | An important parameter in our method is λ in Eq. (4) that determines the impact of relation amongst tasks in the learning process. Another important parameter is the number of selected features. Hence, we study how the performance of our model varies with λ and the number of selected features. Figure 1 shows the performance of our model with different parameter settings which achieves the peak of 84.86% when λ = 0.01 and 1400 features was selected. |