A Deep Modular RNN Approach for Ethos Mining

Authors: Rory Duthie, Katarzyna Budzynska

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

Reproducibility Variable Result LLM Response
Research Type Experimental Annotation of ethotic statements is reliable and its extraction is robust (macro-F1 = 0.83), while annotation of polarity is perfect and its extraction is solid (macro-F1 = 0.84). Table 4 provides +/-ESE results using machine learning classifiers, TF-IDF vectorization and a combination of lexicons. Table 5 shows the results of the ESE/n-ESE classification with a combination of standard machine learning algorithms and RNN and CNN models compared against a baseline classifier using the training set distributions and against the ESE/n-ESE classification from [Duthie et al., 2016].
Researcher Affiliation Academia 1 Centre for Argument Technology, University of Dundee 2 Centre for Argument Technology, Institute of Philosophy and Sociology, Polish Academy of Sciences
Pseudocode No The paper describes a pipeline and a neural network architecture but does not include any pseudocode or algorithm blocks.
Open Source Code No The paper mentions using the keras deep learning library (https://github.com/ fchollet/keras) but does not state that their own implementation code for the DMRNN or ethos mining pipeline is open-source or publicly available.
Open Datasets Yes We construct our publicly available corpus, Ethos Hansard1 (http://arg.tech/Ethos_Hansard1, see Table 1)
Dataset Splits Yes We split our data into 60 training transcripts, of which we use 10% as validation data, and 30 test transcripts to give a wide range of test cases.
Hardware Specification No The paper does not provide specific hardware details such as GPU or CPU models used for running the experiments.
Software Dependencies No The paper mentions using the keras deep learning library, scikit-learn, and the Stanford parser, but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes Table 3: Hyper-parameter values for the DMRNN. Embedding dimension input length: 400, output: 128 Hidden layer dimension 10 POL hidden layer 30 Dropout rate 0.20 LSTM layer 128 Max pooling size 4 Adam α: 0.001, β1: 0.9, β2: 0.999