Segmentation of Tweets with URLs and its Applications to Sentiment Analysis

Authors: Abdullah Aljebreen, Weiyi Meng, Eduard Dragut12480-12488

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present an extensive empirical evaluation of our approach in this section to show (i) the effectiveness of our segmentation algorithm and (ii) its benefit to sentiment analysis on tweets.
Researcher Affiliation Academia 1 Temple University 2 Binghamton University
Pseudocode Yes Algorithm 1: The pseudo code of the main function of our algorithm.
Open Source Code No The paper mentions using a third-party library 'Twitter4J1' and provides its URL, but it does not provide an explicit statement or link to the open-source code for their own developed methodology.
Open Datasets Yes We run our SA experiments on two datasets: ... (2) Sem Eval is a collection of 60k labeled tweets for the Sem Eval tasks (2013-2017) (Rosenthal, Farra, and Nakov 2017).
Dataset Splits No The paper describes sampling methods (simple random and stratified random sampling) to approximate accuracy via manual inspection of 1,000 tweets, but it does not specify traditional training/validation/test dataset splits used for model development or hyperparameter tuning.
Hardware Specification No The paper does not specify any particular hardware (e.g., GPU, CPU models, or cloud instance types) used for running its experiments.
Software Dependencies No The paper mentions using 'Twitter4J1' and 'Stanford Core NLP, BERT' tools, but it does not provide specific version numbers for these or other software dependencies used in their implementation.
Experiment Setup No The paper describes its greedy algorithm and some general steps (e.g., 'similarity threshold'), but it does not provide specific numerical hyperparameters (like learning rates, batch sizes, epochs) or detailed system-level training configurations for their experiments.