Replication Package for "An Empirical Study of the Effectiveness of an Ensemble of Stand-alone Sentiment Detection Tools for Software Engineering Datasets"

Last update: Oct 06, 2022

Related tags

Deep Learning HybridSESentimentTOSEM

Overview

Replication Package for "An Empirical Study of the Effectiveness of an Ensemble of Stand-alone Sentiment Detection Tools for Software Engineering Datasets"

The file "SentiseadCode.zip" contains the source code of Sentisead using Random Forest.
https://www.kaggle.com/giasuddin/hybridsentiment contains the code we used to experiment with advanced language-based pre-trained transformer models (PTM)
The file "Results in Invididual Datasets.docx" shows performance of individual and hybrid tools on the entire datasets as well as the six datasets.
The file "ResultsConsolidatedWithEnsambleAssessment_BERTLabeledFinal.xlsx" contains all the data we used to answer RQ1-RQ8. The following columns are necessary in the above file to understand the data:
1. "File" colmun shows the six studied datasets. This column also shows the folds per file that we used. For example, "DatasetLinJIRA_Cleaned_test_0" denotes the fold 0 in the Lin et al. JIRA dataset, "DatasetLinJIRA_Cleaned_test_1" denotes fold 1, etc.
2. "Sentence" column shows the raw sentence
3. "ManualLabel" column shows the grouth turth sentiment polarity label of the sentence
4. "DsoLabelFullText" shows the polarity label of Opiner on the sentence
5. "POME" shows the polarity label of POME on the sentence
6. "Senti4SD" shows the polarity label of Senti4SD on the sentence after 10-fold cross validation per dataset
7. "SentiCR" shows the polarity label of SentiCR on the sentence after 10-fold cross validation per dataset
8. "SentistrengthSE" shows the polarity label of SentistrengthSE on the sentence
9. "Shannon" shows the entropy score of the sentence based on all texts
10. "ShannonAdjective" shows the entropy score of the sentence based on all adjectives
11. "Ensemble_RF" shows the polarity label of a sentence from Sentiead_B (RQ4) after 10-fold cross validation per dataset
12. 'BERT', 'ROBERTA', 'XLNET', 'ALBERT' show the polarity label of a sentence from stand-alone PTMs (RQ6) after 10-fold cross validation per dataset
13. 'SentiseadBert', 'SentiseadRoberta', 'SentiseadXlnet', 'SentiseadAlbert' show the polarity label of a sentence from Sentiead_{PTM} (RQ7) after 10-fold cross validation per dataset
14. 'SentiseadBertPlus', 'SentiseadRobertaPlus', 'SentiseadXlnetPlus', 'SentiseadAlbertPlus' show the polarity label of a sentence from Sentiead_{PTM}+ (RQ8) after 10-fold cross validation per dataset

Replication Package for "An Empirical Study of the Effectiveness of an Ensemble of Stand-alone Sentiment Detection Tools for Software Engineering Datasets"

Related tags

Overview

Replication Package for "An Empirical Study of the Effectiveness of an Ensemble of Stand-alone Sentiment Detection Tools for Software Engineering Datasets"

Owner

converts nominal survey data into a numerical value based on a dictionary lookup.

Hippocampal segmentation using the UNet network for each axis

Action Segmentation Evaluation

An official implementation of "Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation" (CVPR 2021) in PyTorch.

End-to-end image segmentation kit based on PaddlePaddle.

Neural Caption Generator with Attention

Decentralized Reinforcment Learning: Global Decision-Making via Local Economic Transactions (ICML 2020)

Julia package for contraction of tensor networks, based on the sweep line algorithm outlined in the paper General tensor network decoding of 2D Pauli codes

A implemetation of the LRCN in mxnet

Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

Learning Domain Invariant Representations in Goal-conditioned Block MDPs

Deep Learning Based EDM Subgenre Classification using Mel-Spectrogram and Tempogram Features"

LOFO (Leave One Feature Out) Importance calculates the importances of a set of features based on a metric of choice,

The PyTorch re-implement of a 3D CNN Tracker to extract coronary artery centerlines with state-of-the-art (SOTA) performance. (paper: 'Coronary artery centerline extraction in cardiac CT angiography using a CNN-based orientation classiﬁer')

😮The official implementation of "CoNeRF: Controllable Neural Radiance Fields" 😮

Code for the paper "Attention Approximates Sparse Distributed Memory"

Trading Gym is an open source project for the development of reinforcement learning algorithms in the context of trading.

Library for implementing reservoir computing models (echo state networks) for multivariate time series classification and clustering.

The implementation for "Comprehensive Knowledge Distillation with Causal Intervention".

Pytorch implementation of Supporting Clustering with Contrastive Learning, NAACL 2021