JCLAL: A Java Framework for Active Learning
Authors: Oscar Reyes, Eduardo Pérez, María del Carmen Rodríguez-Hernández, Habib M. Fardoun, Sebastián Ventura
JMLR 2016 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | A configuration file comprises a series of parameters required to run an algorithm. Below, an example of a configuration file is shown, which we call Margin Sampling.cfg. In this example, a 10-fold cross validation evaluation method is used on the data set ecoli located in the folder datasets. For each fold, 5% of the training set is selected to construct the labelled set and the rest of the instances form the unlabelled set. A pool-based sampling scenario with the Margin Sampling strategy is used. The Naive Bayes algorithm is used as a base classifier. (...) After the experiment is run, a summary report which comprises information about the induced classifier and several performance measures is created. |
| Researcher Affiliation | Academia | Oscar Reyes EMAIL Eduardo P erez EMAIL Department of Computer Science University of Holgu ın Holgu ın, Cuba Mar ıa del Carmen Rodr ıguez-Hern andez EMAIL Department of Computer Science and Systems Engineering University of Zaragoza Zaragoza, Spain Habib M. Fardoun EMAIL Department of Information Systems King Abdulaziz University Jeddah, Saudi Arabia Sebasti an Ventura EMAIL Department of Computer Science and Numerical Analysis University of C ordoba C ordoba, Spain Department of Information Systems King Abdulaziz University Jeddah, Saudi Arabia |
| Pseudocode | No | The paper describes the architecture and features of the JCLAL framework, and provides an XML configuration example. It does not include structured pseudocode or algorithm blocks for any methodology. |
| Open Source Code | Yes | JCLAL is an open source software for researchers and end-users to develop AL methods. (...) Open Source. The source code is free and available under the GNU General Public License (GPL). It is hosted at Source Forge, Git Hub, OSSRH repository provided by Sonatype, and Maven Central Repository. |
| Open Datasets | No | The paper mentions using "the data set ecoli" in an example configuration. However, it does not provide a specific link, DOI, or formal citation with author names and year for this dataset, nor does it explicitly state its public availability beyond being used in an example. |
| Dataset Splits | Yes | In this example, a 10-fold cross validation evaluation method is used on the data set ecoli located in the folder datasets. For each fold, 5% of the training set is selected to construct the labelled set and the rest of the instances form the unlabelled set. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments. It focuses on the software framework and its configuration. |
| Software Dependencies | Yes | The software requires Java 1.7, Apache commons logging 1.1, Apache commons collections 3.2, Apache commons configuration 1.5, Apache commons lang 2.4, JFree Chart 1.0, WEKA 3.7, MULAN 1.4 and JUnit 4.10 (for running tests). |
| Experiment Setup | Yes | The paper provides an XML configuration file example (Margin Sampling.cfg) that specifies various experimental settings, including a random seed (seed="9871234"), number of folds (num-folds>10</num-folds>), percentage of data to select (percentage-to-select>5.0</percentage-to-select>), maximum iterations (max-iteration>50</max-iteration>), batch size (batch-size>1</batch-size>), and the base classifier type (weka.classifiers.bayes.Naive Bayes). |