An Enhanced Deep Learning Model for Duplicate Question Detection on Quora Question pairs using Siamese LSTM

Chandra, Mainak; Rodrigues, Andrea; George, Jossy P.

doi:10.1109/icdcece53908.2022.9792906

Cited by 7 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When compared to the FCN model, the LSTM model performed better in terms of accuracy, precision, and recall. It obtained 70.53% train accuracy and 70.49% test accuracy, with a precision of 78.14% and a recall of 27.89% in consistency with [18] study, which used a Siamese LSTM to assess the semantic similarity of two queries in order to improve prediction. The Siamese network is an architecture composed of parallel neural networks, namely LSTM units, for the parallel processing of two questions, with each question passing through an Embedding Layer, an LSTM unit, and then a dense layer.…”

Section: Discussionsupporting

confidence: 77%

See 1 more Smart Citation

Enhancing Question Pairs Identification with Ensemble Learning: Integrating Machine Learning and Deep Learning Models

Tarek,

Noaman,

Kayed

2023

IJACSA

View full text Add to dashboard Cite

The effectiveness of machine learning (ML) and deep learning (DL) models on the Quora question pairs dataset is investigated in this study. ML models, including AdaBoost, reached 73.44% test accuracy, while ensemble learning approaches enhanced outcomes even further, with the Hard-Voting Ensemble achieving 76.13%. DL models, such as FCN, demonstrated test accuracy of 81% with cross validation. These findings contribute to natural language processing by demonstrating the potential of ensemble learning for ML models and the DL models' detailed pattern-capturing capacity.

show abstract

Section: Discussionsupporting

confidence: 77%

“…Chandra and his colleagues [18] provided a technique for detecting duplicate question pairs in their study by dividing the www.ijacsa.thesai.org selected dataset in a 70:30 ratio. A technique known as random splitting was employed.…”

Section: Chandra and Stefanusmentioning

confidence: 99%