Siamese Based Network for Detecting Quora Questions Similarity
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Quora search engine redirects to different discussion pages based on the search terms searched by the user. So, when questions that are semantically similar are searched on Quora, it sometimes redirects a user to different discussion pages even if there exists a page to the dedicated search. In such a case, Semantic Similarity among the questions carries highest weightage. So, for text, using traditional methods for calculating similarity, usually the text is considered as sequence of words and they just count the number of words that occurred in a sentence, on which some distance measures are applied to find the similarity, while missing the semantic level knowledge of the text during calculation. Considering such traditional methods, it will also require a huge training set as well as time to produce an accurate model. But in this Research Paper, Siamese Based Network is used that can train itself on a single example of each text to provide an accurate similarity output.I have used different types of pre-trained word embedding models like word-2-vec and glove to understand the semantics of the question pairs present in Quora Question Pair dataset. This paper introduces a new approach to calculate sentence similarity and gives astronishing results outperforming the current state of art Siamese Based LSTM models. Along with this new approach of using Manhattan LSTM with attention mechanism for similarity calculation, a comparative analysis is performed on the embedded question pairs, among different Siamese based LSTM models like LSTM and Manhattan LSTM, to predict whether the questions are similar or not and get the best model combination for Quora Question Pair
Abstract
Quora search engine redirects to different discussion pages based on the search terms searched by the user. So, when questions that are semantically similar are searched on Quora, it sometimes redirects a user to different discussion pages even if there exists a page to the dedicated search. In such a case, Semantic Similarity among the questions carries highest weightage. So, for text, using traditional methods for calculating similarity, usually the text is considered as sequence of words and they just count the number of words that occurred in a sentence, on which some distance measures are applied to find the similarity, while missing the semantic level knowledge of the text during calculation. Considering such traditional methods, it will also require a huge training set as well as time to produce an accurate model. But in this Research Paper, Siamese Based Network is used that can train itself on a single example of each text to provide an accurate similarity output.I have used different types of pre-trained word embedding models like word-2-vec and glove to understand the semantics of the question pairs present in Quora Question Pair dataset. This paper introduces a new approach to calculate sentence similarity and gives astronishing results outperforming the current state of art Siamese Based LSTM models. Along with this new approach of using Manhattan LSTM with attention mechanism for similarity calculation, a comparative analysis is performed on the embedded question pairs, among different Siamese based LSTM models like LSTM and Manhattan LSTM, to predict whether the questions are similar or not and get the best model combination for Quora Question Pair
참고문헌 (28)
- "Siamese Neural Networks for One-Shot Image Recognition" (2015)[URL]
- "Siamese Network Keras for Image and Text similarity" (2019)[URL]
- "One Shot Learning with Siamese Networks using Keras" (2019)[URL]
- Neculoiu Paul, "Versteegh Maarten and Rotaru Mihai, “Learning Text Similarity with Siamese Recurrent Networks", ~ (2016)
- Ranasinghe, Tharindu Orasan, "Semantic Textual Similarity with Siamese Neural Networks", ~ (2019)
- Thyagarajan Aditya, "Siamese Recurrent Architectures for Learning Sentence Similarity", ~ (2015)
- Gomaa, Wael, "A Survey of Text Similarity Approaches", International journal of Computer Applications, 68 (2013)[DOI]
- Peters Matthew, "Deep Contextualized Word Representations", 2227~2237 (2018)
- Woo HoSung, "Validation of Text Data Preprocessing Using a Neural Network Model", Mathematical Problems in Engineering, pp.1~9 (2020)[DOI]
- Ichida Alexandre, "Measuring Semantic Similarity Between Sentences Using A Siamese Neural Network", 1~7 (2018)
- Imtiaz Zainab, "Duplicate Questions Pair Detection Using Siamese MaLSTM", IEEE Access, pp.1~ (2020)[DOI]
- Zhao Linchang, "Siamese Dense Neural Network for Software Defect Prediction With Small Data", IEEE Access, pp.1~ (2018)[DOI]
- Zhengguang li, "Interactive Self-Attentive Siamese Network for Biomedical Sentence Similarity", IEEE Access, pp.1~ (2020)[DOI]
- Verma Dhruv, "Semantic similarity between short paragraphs using Deep Learning", 1~5 (2020)
- Huang Yangchen, "Person Entity Attribute Extraction Based on Siamese Network", IEEE Access, pp.1~ (2019)[DOI]
- Li Xiang, "Retrieving and ranking short medical questions with two stages neural matching model", ~ (2020)
- Ronran Chirawan, "Effect of Character and Word Features in Bidirectional LSTM-CRF for NER", 613~616 (2020)
- Jiang Keyuan, "Identifying personal health experience tweets with deep neural networks", Conference proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, pp.1174~1177 (2017)
- Huang Zhiheng, "Bidirectional LSTM-CRF Models for Sequence Tagging", ~ (2015)
- Liang Guoxi, "Automated Essay Scoring: A Siamese Bidirectional LSTM Neural Network Architecture", Symmetry, 10, pp.682~ (2018)[DOI]
- Linhares Pontes, "Predicting the Semantic Textual Similarity with Siamese CNN and LSTM", ~ (2018)
- Zhu Wenhao, "Dependency-based Siamese long short-term memory network for learning sentence representations", PLOS ONE, 13, pp.e0193919~ (2018)[DOI]
- Bao Wei, "Attentive Siamese LSTM Network for Semantic Textual Similarity Measure", 312~317 (2018)
- Zhang Lingxi, "Siamese-Based BiLSTM Network for Scratch Source Code Similarity Measuring", 1800~1805 (2020)
- "Do More With Less Data! —One-shot Learning with Siamese Neural Network" (2020)[URL]
- Neculoiu Paul, "Learning Text Similarity with Siamese Recurrent Networks", ~ (2016)
- Imtiaz Zainab, "Duplicate Questions Pair Detection Using Siamese MaLSTM", IEEE Access, pp.1~ (2020)[DOI]
- 김대현, "Application of Deep Neural Network Model for Automated Intelligent Excavator", 아시아태평양융합연구교류논문지, 6(4), pp.13~22 (2020), 사단법인 한국융합기술연구학회[DOI]