scholarly journals A Systematic Study on Sentiment Analysis based on Text Mining and Deep learning for Predictions in Stock Market Trends Through Social and News Media Data

2021 ◽  
Vol 9 (10) ◽  
pp. 1589-1593
Author(s):  
Suvigya Jain
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Stock Market ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
News Media ◽  
Market Sentiment ◽  
Market Trends

Abstract: Stock Market has always been one of the most active fields of research, many companies and organizations have focused their research in trying to find better ways to predict market trends. The stock market has been the instrument to measure the performance of a company and many have tried to develop methods that reduce risk for the investors. Since, the implementation of concepts like Deep Learning and Natural Language Processing has been made possible due to modern computing there has been a revolution in forecasting market trends. Also, the democratization of knowledge related to companies made possible due to the internet has provided the stake holders a means to learn about assets they choose to invest in through news media and social media also stock trading has become easier due to apps like robin hood etc. Every company now a days has some kind of social media presence or is usually reported by news media. This presence can lead to the growth of the companies by creating positive sentiment and also many losses by creating negative sentiments due to some public events. Our goal in this paper is to study the influence of news media and social media on market trends using sentiment analysis. Keywords: Deep Learning, Natural Language Processing, Stock Market, Sentiment analysis

Deep Learning Techniques for Polarity Classification in Multimodal Sentiment Analysis

2018 ◽  
Vol 17 (03) ◽  
pp. 883-910 ◽  
Author(s):  
P. D. Mahendhiran ◽  
S. Kannimuthu
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Relevant Information ◽  
Neural Net ◽  
Video Information ◽  
Automatic Feature Extraction ◽  
Multimodal Sentiment Analysis

Contemporary research in Multimodal Sentiment Analysis (MSA) using deep learning is becoming popular in Natural Language Processing. Enormous amount of data are obtainable from social media such as Facebook, WhatsApp, YouTube, Twitter and microblogs every day. In order to deal with these large multimodal data, it is difficult to identify the relevant information from social media websites. Hence, there is a need to improve an intellectual MSA. Here, Deep Learning is used to improve the understanding and performance of MSA better. Deep Learning delivers automatic feature extraction and supports to achieve the best performance to enhance the combined model that integrates Linguistic, Acoustic and Video information extraction method. This paper focuses on the various techniques used for classifying the given portion of natural language text, audio and video according to the thoughts, feelings or opinions expressed in it, i.e., whether the general attitude is Neutral, Positive or Negative. From the results, it is perceived that Deep Learning classification algorithm gives better results compared to other machine learning classifiers such as KNN, Naive Bayes, Random Forest, Random Tree and Neural Net model. The proposed MSA in deep learning is to identify sentiment in web videos which conduct the poof-of-concept experiments that proved, in preliminary experiments using the ICT-YouTube dataset, our proposed multimodal system achieves an accuracy of 96.07%.

scholarly journals An Exploration of Impact of COVID 19 on mental health -Analysis of tweets using Natural Language Processing techniques

2020 ◽  
Author(s):  
Sohini Sengupta ◽  
Sareeta Mugde ◽  
Garima Sharma
Keyword(s):  
Mental Health ◽  
Social Media ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Gold Mine ◽  
Social Media Platforms ◽  
Processing Techniques

Twitter is one of the world's biggest social media platforms for hosting abundant number of user-generated posts. It is considered as a gold mine of data. Majority of the tweets are public and thereby pullable unlike other social media platforms. In this paper we are analyzing the topics related to mental health that are recently (June, 2020) been discussed on Twitter. Also amidst the on-going pandemic, we are going to find out if covid-19 emerges as one of the factors impacting mental health. Further we are going to do an overall sentiment analysis to better understand the emotions of users.

Deep Learning Approaches for Textual Sentiment Analysis

2020 ◽  
pp. 171-182 ◽  
Author(s):  
Tamanna Sharma ◽  
Anu Bajaj ◽  
Om Prakash Sangwan
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Machine Learning Techniques ◽  
Computational Technique ◽  
Learning Techniques

Sentiment analysis is computational measurement of attitude, opinions, and emotions (like positive/negative) with the help of text mining and natural language processing of words and phrases. Incorporation of machine learning techniques with natural language processing helps in analysing and predicting the sentiments in more precise manner. But sometimes, machine learning techniques are incapable in predicting sentiments due to unavailability of labelled data. To overcome this problem, an advanced computational technique called deep learning comes into play. This chapter highlights latest studies regarding use of deep learning techniques like convolutional neural network, recurrent neural network, etc. in sentiment analysis.

Coronavirus Pandemic (COVID-19)

2021 ◽  
Vol 17 (2) ◽  
pp. 1-21
Author(s):  
Jalal S. Alowibdi ◽  
Abdulrahman A. Alshdadi ◽  
Ali Daud ◽  
Mohamed M. Dessouky ◽  
Essa Ali Alhazmi
Keyword(s):  
Social Media ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Experimental Results ◽  
Mixed Emotions ◽  
Social Media Platforms

People are afraid about COVID-19 and are actively talking about it on social media platforms such as Twitter. People are showing their emotions openly in their tweets on Twitter. It's very important to perform sentiment analysis on these tweets for finding COVID-19's impact on people's lives. Natural language processing, textual processing, computational linguists, and biometrics are applied to perform sentiment analysis to identify and extract the emotions. In this work, sentiment analysis is carried out on a large Twitter dataset of English tweets. Ten emotional themes are investigated. Experimental results show that COVID-19 has spread fear/anxiety, gratitude, happiness and hope, and other mixed emotions among people for different reasons. Specifically, it is observed that positive news from top officials like Trump of chloroquine as cure to COVID-19 has suddenly lowered fear in sentiment, and happiness, gratitude, and hope started to rise. But, once FDA said, chloroquine is not effective cure, fear again started to rise.

High accuracy offering attention mechanisms based deep learning approach using CNN/bi-LSTM for sentiment analysis

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Venkateswara Rao Kota ◽  
Shyamala Devi Munisamy
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Attention Mechanism ◽  
Supervised Machine Learning ◽  
Method Performance ◽  
Content Type

PurposeNeural network (NN)-based deep learning (DL) approach is considered for sentiment analysis (SA) by incorporating convolutional neural network (CNN), bi-directional long short-term memory (Bi-LSTM) and attention methods. Unlike the conventional supervised machine learning natural language processing algorithms, the authors have used unsupervised deep learning algorithms.Design/methodology/approachThe method presented for sentiment analysis is designed using CNN, Bi-LSTM and the attention mechanism. Word2vec word embedding is used for natural language processing (NLP). The discussed approach is designed for sentence-level SA which consists of one embedding layer, two convolutional layers with max-pooling, one LSTM layer and two fully connected (FC) layers. Overall the system training time is 30 min.FindingsThe method performance is analyzed using metrics like precision, recall, F1 score, and accuracy. CNN is helped to reduce the complexity and Bi-LSTM is helped to process the long sequence input text.Originality/valueThe attention mechanism is adopted to decide the significance of every hidden state and give a weighted sum of all the features fed as input.

scholarly journals Sentiment Analysis for Depression Based on Social Media Post by Using Natural Language Processing

2021 ◽  
pp. 134-139
Author(s):  
G. Neelavathi ◽  
D. Sowmiya ◽  
C. Sharmila ◽  
J. Vaishnavi
Keyword(s):  
Social Media ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Rich Source ◽  
Research Center ◽  
Depression Symptoms ◽  
Linguistic Markers

Presently Research Center expresses that, 72% of public uses some sort of social media. More than 300 million individual experiences the depression and despondency, just a small amount of them get sufficient treatment. Discouragement is the main source of incapacity worldwide and almost 800,000 individuals consistently loss their life because of suicide. Suicide is the subsequent driving reason for death among teenagers. Our idea is to suggest solution for this problem. Social Media gives an extraordinary chance to change early depressions, especially in youngsters. Consistently, around 6,000 Tweets are tweeted per second, 350,000 tweets per minute, 500 million tweets each day and around 200 billion tweets each year. By using this rich source of data and information, can efficient model which provides report of person’s depression symptoms will be designed. In this model an algorithm that can examine Tweets Expressing self-assessed negative features by analyzing linguistic markers in social media posts.

scholarly journals A Review on the Detection of Offensive Content in Social Media Platforms

2021 ◽  
Vol 6 (1) ◽  
Author(s):  
Solomon Akinboro ◽  
Oluwadamilola Adebusoye ◽  
Akintoye Onamade
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Hybrid Methods ◽  
Future Research ◽  
Learning Approaches ◽  
Hybrid Approaches ◽  
Use Of Social Media

Offensive content refers to messages which are socially unacceptable including vulgar or derogatory messages. As the use of social media increases worldwide, social media administrators are faced with the challenges of tackling the inclusion of offensive content, to ensure clean and non-abusive or offensive conversations on the platforms they provide.  This work organizes and describes techniques used for the automated detection of offensive languages in social media content in recent times, providing a structured overview of previous approaches, including algorithms, methods and main features used.   Selection was from peer-reviewed articles on Google scholar. Search terms include: Profane words, natural language processing, multilingual context, hybrid methods for detecting profane words and deep learning approach for detecting profane words. Exclusions were made based on some criteria. Initial search returned 203 of which only 40 studies met the inclusion criteria; 6 were on natural language processing, 6 studies were on Deep learning approaches, 5 reports analysed hybrid approaches, multi-level classification/multi-lingual classification appear in 13 reports while 10 reports were on other related methods.The limitations of previous efforts to tackle the challenges with regards to the detection of offensive contents are highlighted to aid future research in this area.  Keywords— algorithm, offensive content, profane words, social media, texts

scholarly journals Deep learning for pharmacovigilance: recurrent neural network architectures for labeling adverse drug reactions in Twitter posts

2017 ◽  
Vol 24 (4) ◽  
pp. 813-821 ◽  
Author(s):  
Anne Cocos ◽  
Alexander G Fiks ◽  
Aaron J Masino
Keyword(s):  
Neural Network ◽  
Social Media ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Recurrent Neural Network ◽  
State Of The Art ◽  
Detection Performance ◽  
Word Embeddings

Abstract Objective Social media is an important pharmacovigilance data source for adverse drug reaction (ADR) identification. Human review of social media data is infeasible due to data quantity, thus natural language processing techniques are necessary. Social media includes informal vocabulary and irregular grammar, which challenge natural language processing methods. Our objective is to develop a scalable, deep-learning approach that exceeds state-of-the-art ADR detection performance in social media. Materials and Methods We developed a recurrent neural network (RNN) model that labels words in an input sequence with ADR membership tags. The only input features are word-embedding vectors, which can be formed through task-independent pretraining or during ADR detection training. Results Our best-performing RNN model used pretrained word embeddings created from a large, non–domain-specific Twitter dataset. It achieved an approximate match F-measure of 0.755 for ADR identification on the dataset, compared to 0.631 for a baseline lexicon system and 0.65 for the state-of-the-art conditional random field model. Feature analysis indicated that semantic information in pretrained word embeddings boosted sensitivity and, combined with contextual awareness captured in the RNN, precision. Discussion Our model required no task-specific feature engineering, suggesting generalizability to additional sequence-labeling tasks. Learning curve analysis showed that our model reached optimal performance with fewer training examples than the other models. Conclusions ADR detection performance in social media is significantly improved by using a contextually aware model and word embeddings formed from large, unlabeled datasets. The approach reduces manual data-labeling requirements and is scalable to large social media datasets.

Sign in / Sign up

Export Citation Format

Share Document