Relational One-Class Classification: A Non-Parametric Approach

Authors: Tushar Khot, Sriraam Natarajan, Jude Shavlik

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4 Experiments We evaluate the following different algorithms: [...] We use area under the Precision-Recall curve (AUC-PR) and accuracy to compare our approaches.
Researcher Affiliation Academia Tushar Khot and Sriraam Natarajan and Jude Shavlik University of Wisconsin-Madison, {tushar,shavlik}@cs.wisc.edu Indiana University, natarasr@indiana.edu
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes UW-CSE dataset (Richardson and Domingos 2006) is a standard SRL dataset developed at University of Washington. [...] IMDB is another popular SRL dataset (Mihalkova and Mooney 2007) containing information about actors, directors and movies obtained from imdb.com. [...] we use the Heart dataset which is a multivariate data set with 13 attributes.
Dataset Splits Yes We pick 20%, 40% and 60% of the labeled examples for training and use all the positives for testing. The results for five-fold cross-validation are shown in Table 2 and Table 3.
Hardware Specification No The paper does not provide specific hardware details (like GPU/CPU models or types) used for running its experiments.
Software Dependencies No The paper mentions using 'the Stanford NLP toolkit' but does not specify a version. It also mentions 'SVM' and 'TILDE' but without specific library or software version numbers.
Experiment Setup Yes The parameter λ (set to 0.5 in experiments) ensures that the distance value decreases gradually as the depth increases. [...] For the weight learning step, we use a step length of η = 0.001.