Slogans Are Not Forever: Adapting Linguistic Expressions to the News

Authors: Lorenzo Gatti, Gözde özbal, Marco Guerini, Oliviero Stock, Carlo Strapparava

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The evaluation that we carried out confirms the effectiveness of our approach for evoking a novel concept by modifying these well known expressions.
Researcher Affiliation Academia Lorenzo Gatti FBK-irst Trento, Italy l.gatti@fbk.eu Gozde Ozbal FBK-irst Trento, Italy gozbalde@gmail.com Marco Guerini Trento RISE Trento, Italy marco.guerini@trentorise.eu Oliviero Stock FBK-irst Trento, Italy stock@fbk.eu Carlo Strapparava FBK-irst Trento, Italy strappa@fbk.eu
Pseudocode No The paper describes the system components and their processes but does not include any formal pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any links to open-source code for the methodology described, nor does it state that the code is publicly available.
Open Datasets Yes Each probability of a lemma is calculated as the number of headlines that it appears divided by the total number of headlines occurring in the corpus. Then, we tokenize and Po S-tag the daily snippets with Stanford Parser [Klein and Manning, 2003] and lemmatize them with Word Net. ... LDC Giga Word 5th Edition corpus4 (http://www.ldc.upenn.edu/Catalog/catalog Entry.jsp?catalog Id= LDC2011T07)
Dataset Splits No The paper describes using a similarity threshold ('The similarity filter was set to cos > 0.6. Threshold values were determined empirically.') and a crowdsourced evaluation on '16 pairs of sentences for each strategy'. However, it does not specify explicit training/validation/test splits for model development or evaluation in the traditional machine learning sense.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU models, CPU types, memory) used to run the experiments.
Software Dependencies No The paper mentions several tools and resources like 'Stanford Parser [Klein and Manning, 2003]', 'Word Net', 'Stanford Named Entity Recognizer [Finkel et al., 2005]', 'Free Base [Bollacker et al., 2008]', 'skip-gram model [Mikolov et al., 2013]', and 'Morpho Pro [Pianta et al., 2008]'. However, it does not specify version numbers for these software components or any programming languages/libraries used for implementation.
Experiment Setup Yes The system takes up to 10 most similar articles among the ones with a similarity value greater than a given threshold9. The similarity filter was set to cos > 0.6. Threshold values were determined empirically.