Effective Broad-Coverage Deep Parsing

Authors: James Allen, Omid Bahkshandeh, William de Beaumont, Lucian Galescu, Choh Man Teng

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate these customizations with some ablation experiments.
Researcher Affiliation Academia IHMC, 40 S. Alcaniz St, Pensacola, FL 32501 {jallen, omidb, wbeaumont, lgalescu, cmteng}@ihmc.us
Pseudocode No The paper does not contain any sections or figures explicitly labeled as 'Pseudocode' or 'Algorithm'.
Open Source Code No The paper provides links to the general TRIPS parser system and its lexicon/ontology (e.g., http://trips.ihmc.us/parser and www.cs.rochester.edu/research/trips/lexicon/browse-ont-lex.html), but does not explicitly state that the source code for the specific methodology or customizations described in this paper is available.
Open Datasets No While the experimental data is extracted from Pub Med Central, the specific set of 60 sentences and their gold annotations are not made publicly available with a link, DOI, or formal citation for the dataset as used in the experiment.
Dataset Splits No The paper mentions using 60 sentences and constructing a 'test set', but does not specify any train/validation/test splits, percentages, or absolute counts for dataset partitioning.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions several software tools used (e.g., 'Stanford part-of-speech tagger', 'Enju parser'), but does not provide specific version numbers for these software components, which is necessary for reproducibility.
Experiment Setup Yes All the ablation tests in Figure 4 were performed with a bracket crossing penalty of 0.99 and a boosting parameter of 0.2.