Convolution Kernels for Discriminative Learning from Streaming Text
Authors: Michal Lukasik, Trevor Cohn
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our method outperforms competitive baselines in three synthetic and two real datasets, rumour frequency modeling and popularity prediction tasks. |
| Researcher Affiliation | Academia | Michal Lukasik Computer Science The University of Sheffield m.lukasik@shef.ac.uk Trevor Cohn Computing and Information Systems The University of Melbourne t.cohn@unimelb.edu.au |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper states: 'The work was implemented using the GPy toolkit (The GPy authors 2015).', and provides a URL to the GPy framework. This refers to a third-party tool used, not the authors' own implementation code for the described methodology. |
| Open Datasets | Yes | We conduct experiments on two social media datasets. The first dataset consists of 114 popular rumours collected in August 2014 during the Ferguson unrest and is composed of 4098 tweets. This dataset has been used by Lukasik, Cohn, and Bontcheva (2015) for rumour dynamics modeling. (...) We also consider a second dataset consisting of tweets collected October 2014 during Ottawa shootings and in August 2014 during the Ferguson unrest (Zubiaga et al. 2015). |
| Dataset Splits | Yes | In case of synthetic experiments, we ran 55 experiments of each setting, taking 80% of events for training and 20% for testing. (...) Evaluation is run in a leave one out manner, where the prediction is made for each rumour, using the first hour of tweets from the rumour as well as (depending on a method) full two hours of tweets for the remaining 113 rumours. (...) evaluate the predictive accuracy with 5-fold cross validation |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types, or memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper states: 'The work was implemented using the GPy toolkit (The GPy authors 2015).'. While it names a software toolkit, it does not provide specific version numbers for GPy or other key software components, which is required for reproducibility. |
| Experiment Setup | No | The paper describes the general framework and theoretical underpinnings (e.g., Gaussian Processes, kernel choices) and how hyperparameters are learned ('learning the hyperparameter values (σ1, σ2, l)'), but it does not specify concrete numerical values for hyperparameters or other explicit training configurations (e.g., batch size, learning rates, epochs) used in the experiments. |