Parallel Symmetric Class Expression Learning
Authors: An C. Tran, Jens Dietrich, Hans W. Guesgen, Stephen Marsland
JMLR 2017 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate that this algorithm provides significantly better results (in terms of metrics such as accuracy, search space covered, and learning time) than standard approaches on some typical data sets. Further, the approach has the added benefit that it can be parallelised relatively simply, leading to much faster exploration of the search tree on modern computers. |
| Researcher Affiliation | Academia | An C. Tran EMAIL College of Information and Communication Technology Can Tho University, Vietnam Jens Dietrich EMAIL Hans W. Guesgen EMAIL Stephen Marsland EMAIL School of Engineering and Advanced Technology Massey University, New Zealand. |
| Pseudocode | Yes | Algorithm 1: Symmetric Class Expression Learning Algorithm SPa CEL(K, E+, E , ε) Algorithm 2: Specialisation algorithm Specialise(C, K, E+, E ) Algorithm 3: Combination algorithm Combine(C, cpdefs, E ) |
| Open Source Code | No | The implementation of CELOE that we used is publicly available at https://github.com/tcanvn/SPar CEL/. ... The data set is available in the SPa CEL repository6. |
| Open Datasets | Yes | Moral This data set was first introduced by Wogulis (1994) and is available at the University of California Irvine (UCI) machine learning repository2 (Frank and Asuncion, 2010). ... Poker A description of 10 classes of poker hands ... available at the UCI repository3. ... Family A set of data sets ... In our evaluation, we use the data set distributed with the DL-Learner package (which can be found in the DL-Learner repository4). ... Carcino Genesis This data set ... was transformed into the OWL ontology format and used to evaluate the description logic learning algorithms in the DL-Learner framework (Lehmann, 2010)5. ... MUSE Dataset We employ use case 1 of a set of context-aware smart home simulations by Lyons et al. (2010) ... The data set is available in the SPa CEL repository6. ... ILPD Dataset ... This data set is also available in the UCI repository7. ... MUBus Dataset ... This data set is available in the SPa CEL repository; |
| Dataset Splits | Yes | In order to evaluate the Symmetric Class Expression Learning approach of SPa CEL, we compared it to CELOE and Par CEL using 10-fold cross validation with 15 data sets. |
| Hardware Specification | No | The paper mentions "modern computers", "multi-core processors or cloud computing platforms" and "multiple workers" but does not provide specific details such as GPU/CPU models, memory, or clock speeds used for the experiments. |
| Software Dependencies | No | The paper mentions several software systems and frameworks like CELOE, DL-Learner, PVM, Apache Hadoop, but does not specify exact version numbers for any of them in relation to the experimental setup. |
| Experiment Setup | Yes | Default values for the weightings of the above factors were chosen based on experimental investigations and the work by Lehmann and Hitzler (2010). We chose α = 0.2, β = 0.01, γ = 0.05. ... In order to evaluate the Symmetric Class Expression Learning approach of SPa CEL, we compared it to CELOE and Par CEL using 10-fold cross validation with 15 data sets. ... all algorithms in the experiment can find solutions on the training set without timing out in the 10 minutes we allowed. ... the experiment was rerun for 3 hours |