A Sentiment Based Non-Factoid Question-Answering Framework

Ye, Qiaofei; Misra, Kanishka; Devarapalli, Hemanth; Rayz, Julia Taylor

doi:10.1109/smc.2019.8913898

Cited by 5 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…-The list of intensifiers has been made using the list of general intensifiers [10,25]. The negators are selected based on Kiritchenko and Saif research [35,75].…”

Section: Resultsmentioning

confidence: 99%

“…Sentiment analysis contains various processes such as subjectivity and polarity detection [4,77], emotion estimation [79], answering to the emotional questions [53], detecting spam comments [75], question answering [70], crime detection [38], sarcasm/irony detection [43], summarizing opinions [44] and many other subjects. Much research are currently conducting to extract user's comments from documents [15], sentences [57], or aspect-based [55].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deterministic solution of algebraic equations in sentiment analysis

Zahedi

Basiri

2023

Multimed Tools Appl

View full text Add to dashboard Cite

Text mining methods usually use statistical information to solve text and language-independent procedures. Text mining methods such as polarity detection based on stochastic patterns and rules need many samples to train. On the other hand, deterministic and non-probabilistic methods are easy to solve and faster than other methods but are not efficient in NLP data. In this article, a fast and efficient deterministic method for solving the problems is proposed. In the proposed method firstly we transform text and labels into a set of equations. In the second step, a mathematical solution of ill-posed equations known as Tikhonov regularization was used as a deterministic and non-probabilistic way including additional assumptions, such as smoothness of solution to assign a weight that can reflect the semantic information of each sentimental word. We confirmed the efficiency of the proposed method in the SemEval-2013 competition, ESWC Database and Taboada database as three different cases. We observed improvement of our method over negative polarity due to our proposed mathematical step. Moreover, we demonstrated the effectiveness of our proposed method over the most common and traditional machine learning, stochastic and fuzzy methods.

show abstract

“…-The list of intensifiers has been made using the list of general intensifiers [10,25]. The negators are selected based on Kiritchenko and Saif research [35,75].…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deterministic solution of algebraic equations in sentiment analysis

Zahedi

Basiri

2023

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Somasundaran et al (2007) propose to use attitude as a part of the feature set to improve QA performance. Similar ideas have been explored in why-QA (Oh et al, 2012), community QA (Elalfy et al, 2015;Eskandari et al, 2015), opinion QA (Ku et al, 2008;Pang and Ngo, 2015), how-QA (Ye et al, 2019), and yes/no QA (Sarrouti and El Alaoui, 2017). Inversely, QA-style sentiment classification has also been investigated (Shen et al, 2018).…”

Section: Sentiment-enhanced Qamentioning

confidence: 91%

“…The yes/no QA task can be intuitively formatted as a binary classification problem, where the input should include the question and the context, and the answer is either yes or no. Feature-based approaches (Somasundaran et al, 2007;Oh et al, 2012) have been investigated over a decade ago, and recently proposed solutions mostly employ deep neural structures (Ye et al, 2019), along with transfer learning (Jin et al, 2019) to optimize the usage of domain knowledge. Pre-trained models, such as ELMo (Peters et al, 2018), ELECTRA (Clark et al, 2020), BERT (Devlin et al, 2018), and its variants (Jin et al, 2019;Liu et al, 2019), have shown superior performance and refreshed the SOTA records.…”

Section: Yes/no Qamentioning

confidence: 99%

SentiMedQAer: A Transfer Learning-Based Sentiment-Aware Model for Biomedical Question Answering

Xiang

Chen²,

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

Recent advances have witnessed a trending application of transfer learning in a broad spectrum of natural language processing (NLP) tasks, including question answering (QA). Transfer learning allows a model to inherit domain knowledge obtained from an existing model that has been sufficiently pre-trained. In the biomedical field, most QA datasets are limited by insufficient training examples and the presence of factoid questions. This study proposes a transfer learning-based sentiment-aware model, named SentiMedQAer, for biomedical QA. The proposed method consists of a learning pipeline that utilizes BioBERT to encode text tokens with contextual and domain-specific embeddings, fine-tunes Text-to-Text Transfer Transformer (T5), and RoBERTa models to integrate sentiment information into the model, and trains an XGBoost classifier to output a confidence score to determine the final answer to the question. We validate SentiMedQAer on PubMedQA, a biomedical QA dataset with reasoning-required yes/no questions. Results show that our method outperforms the SOTA by 15.83% and a single human annotator by 5.91%.

show abstract

ADA: An Acronym-Based Data Augmentation Method for Low-Resource Text Classification

Chen

Xue-shan

et al. 2022

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

A Sentiment Based Non-Factoid Question-Answering Framework

Cited by 5 publications

References 5 publications

Deterministic solution of algebraic equations in sentiment analysis

Deterministic solution of algebraic equations in sentiment analysis

SentiMedQAer: A Transfer Learning-Based Sentiment-Aware Model for Biomedical Question Answering

ADA: An Acronym-Based Data Augmentation Method for Low-Resource Text Classification

Contact Info

Product

Resources

About