User Intent Identification from Online Discussions Using a Joint Aspect-Action Topic Model

Authors: Ghasem Heyrani Nobari, Chua Tat-Seng

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

Reproducibility Variable Result LLM Response
Research Type Experimental In order to demonstrate the effectiveness of our approach, we empirically compare our model against the state of the art methods on a large-scale discussion dataset crawled from the apple discussion forum with over 3.3 million user posts from 340k discussion threads. We carry out extensive evaluation of our joint model on a large scale discussion dataset crawled from apple discussion forum.
Researcher Affiliation Academia Ghasem Heyrani-Nobari, Tat-Seng Chua National University of Singapore {ghasem,dcscts}@nus.edu.sg
Pseudocode Yes Algorithm 1 shows the detail of extracting aspect and action terms from a sequence of user posts. [...] Algorithm 2 Relationship Graph Generation
Open Source Code No The paper does not provide any links or explicit statements about the release of source code for the described methodology.
Open Datasets No The paper states, "We setup our dataset using the data crawled from Apple discussions1." and provides a URL "1http://discussions.apple.com". While the source is public, the specific curated dataset of "3.3 million user posts from 340k discussion threads" used in their experiments is not made publicly available with a direct link, DOI, or formal citation for the dataset itself.
Dataset Splits No The paper mentions using a "testing set" and categorizes the dataset by topic, but it does not provide specific percentages or counts for training, validation, and test splits to reproduce the data partitioning.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU, CPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions using a "POS Tagger" and "shallow parsing methods" but does not specify any software libraries or tools with their version numbers.
Experiment Setup No The paper mentions conducting "experiments with different settings such as the number of topics or iterations" but does not provide the specific values for these or other hyperparameters like learning rates or optimizer settings.