Assessing the Severity of Health States based on Social Media Posts

Yadav, Shweta; Sain, Joy Prakash; Sheth, Amit P.; Ekbal, Asif; Saha, Sriparna; Bhattacharyya, Pushpak

doi:10.1109/icpr48806.2021.9411980

Cited by 4 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…,vn, -0.875, -0.688, -0.6, -0.79, -0.631, -0.804, 1]. According to the size of refined word embedding, the model creates a k-unit hidden layer (k is defined as 100 in this study) with a tanh activation function described in (2).…”

Section: Generating Meta Embedding Matrixmentioning

confidence: 99%

“…CNN has been reported as one of the widely used classifiers in sentiment classification and recent works focused on sentiment analysis of drug reviews also applied CNN for classification purposes. In this model, we adopted the architecture described in [2]. The input layer accepts each drug review as the input to the classification model.…”

Section: Drug Review Sentiment Classification Using Cnnmentioning

confidence: 99%

“…I am with mood swings frequently." Here, "mood swings" as such do not bring any negative meaning in general but in the context of drug reviews, it represents the adverse drug reaction (ADR) which is strongly related to implicitly defined negative sentiment [2]. In such cases, domain knowledge is critical in solving sentiment analysis problems in the medical domain.…”

Section: Introductionmentioning

confidence: 99%

“…Int J Artif Intell ISSN: 2252-8938  Deep learning models have reported promising results on natural language processing (NLP) tasks including sentiment analysis and the popularity of pre-trained word embedding models in the feature extraction process. The use of word embedding in the deep learning models for medical sentiment analysis has been reported recently in [2]- [6]. Word embedding refers to the dense vector representation of words inspired by the distributional hypothesis [7] "The words that occur in the same context have similar meanings".…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A novel meta-embedding technique for drug reviews sentiment analysis

Khine

Wettayaprasit

Duangsuwan

2023

IJ-AI

View full text Add to dashboard Cite

<p>Traditional word embedding models have been used in the feature extraction process of deep learning models for sentiment analysis. However, these models ignore the sentiment properties of words while maintaining the contextual relationships and have inadequate representation for domainspecific words. This paper proposes a method to develop a meta embedding model by exploiting domain sentiment polarity and adverse drug reaction (ADR) features to render word embedding models more suitable for medical sentiment analysis. The proposed lexicon is developed from the medical blogs corpus. The polarity scores of the existing lexicons are adjusted to assign new polarity score to each word. The neural network model utilizes sentiment lexicons and ADR in learning refined word embedding. The refined embedding obtained from the proposed approach is concatenated with original word vectors, lexicon vectors, and ADR feature to form a meta-embedding model which maintains both contextual and sentimental properties. The final meta-embedding acts as a feature extractor to assess the effectiveness of the model in drug reviews sentiment analysis. The experiments are conducted on global vectors (GloVE) and skip-gram word2vector (Word2Vec) models. The empirical results demonstrate the proposed meta-embedding model outperforms traditional word embedding in different performance measures.</p>

show abstract

Section: Generating Meta Embedding Matrixmentioning

confidence: 99%

Section: Drug Review Sentiment Classification Using Cnnmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A novel meta-embedding technique for drug reviews sentiment analysis

Khine

Wettayaprasit

Duangsuwan

2023

IJ-AI

View full text Add to dashboard Cite

show abstract

“…Blogs, Online Systems, Interaction Platforms, etc.) keep dominating the interaction, business, entertainment, political, social and academic activities of people worldwide [1]. In the olden days, it had been challenging to accumulate the opinion of people due to the lack of centralised platforms that gather people and give them the privilege to divulge their beliefs liberally [2].…”

Section: Introductionmentioning

confidence: 99%

Sentiments Analysis Study for the Adaptation of Online Medical Forums

Hassan

Sani

Balarabe

2023

AJRCoS

View full text Add to dashboard Cite

Online medical forums allow users to research medical treatments or conditions and gain support from other users dealing with similar issues. These forums have become increasingly popular over the past decade, helping connect medical patients and professionals from various backgrounds and creating a supportive online community. This paper evaluates the adaptation of online medical forums in Nigeria, to analyse the opinions of Nigerian citizens in using the medical system. In this research, a Tweepy API (python library/ module that contains the required object and functions for managing the Twitter data), textblob (python library for processing textual data), and matplotlib modules (for creating statistical charts) were used to extract related tweets from the Twitter. The project involves steps like creating a Twitter developer account, which gives the privilege to create a Twitter application and has keys for accessing online resources. The analysis begins by searching for the data, storing it, filtering it and then returning the sentiment analysis to review the positive, neutral, and negative tweets. The output of this project returns a table and scatter graph that displays the Subjectivity and polarity of the opinions of Nigerians on the adaptation of online medical forums. Similarly, a bar chart is obtained that shows the positive tweets, the negative tweets and the neutral regarding online medical forums.

show abstract