Cause-Effect Association between Event Pairs in Event Datasets

Authors: Debarun Bhattacharjya, Tian Gao, Nicholas Mattei, Dharmashankar Subramanian

IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct an experimental investigation with synthetic data and two real-world event datasets, where we evaluate and compare our proposed scores using assessments from human raters as ground truth.
Researcher Affiliation Collaboration 1 Research AI, IBM T. J. Watson Research Center 2 Department of Computer Science, Tulane University
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not include any explicit statement about releasing source code or provide a link to a code repository.
Open Datasets Yes We conduct an experimental investigation with synthetic data and two real-world event datasets: 1) a diabetes dataset [Frank and Asuncion, 2010; Acharya, 2014], and 2) the ICEWS political event dataset [O Brien, 2010] a relational (dyadic) event dataset where events are interactions between two actors.
Dataset Splits Yes We split the dataset into equal-sized training/test sets, determine a method s optimal hyper-parameter setting on the training set, and then compute the Hits@K on the test set using this hyper-parameter setting.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers.
Experiment Setup Yes We ran experiments over a sweep of the hyperparameters: α {0, 0.5, 1, 2, 5}, λ {0, 0.25, 0.5, 0.75, 1} for NSTE, γ {0.001, 0.005, 0.01, 0.05, 0.1} for ECDE, g = {avg, max, min} for CIRM and window w = {7, 15, 30} days for all models, using support s = 10. [...] α {0, 1, 5}, λ {0, 0.5, 1} for NSTE, γ {0.001, 0.01, 0.1} for ECDE, g = {avg, max, min} for CIRM and window w = {0.1, 0.3, 0.5, 1} days for all models.