scholarly journals Sarcasm Analysis Using Conversation Context

2018 ◽  
Vol 44 (4) ◽  
pp. 755-792 ◽  
Author(s):  
Debanjan Ghosh ◽  
Alexander R. Fabbri ◽  
Smaranda Muresan
Keyword(s):  
Social Media ◽  
Computational Models ◽  
Human Performance ◽  
Short Term Memory ◽  
Data Sets ◽  
Discussion Forums ◽  
Sentence Level ◽  
Social Media Platforms ◽  
Gold Label ◽  
Lstm Network

Computational models for sarcasm detection have often relied on the content of utterances in isolation. However, the speaker’s sarcastic intent is not always apparent without additional context. Focusing on social media discussions, we investigate three issues: (1) does modeling conversation context help in sarcasm detection? (2) can we identify what part of conversation context triggered the sarcastic reply? and (3) given a sarcastic post that contains multiple sentences, can we identify the specific sentence that is sarcastic? To address the first issue, we investigate several types of Long Short-Term Memory (LSTM) networks that can model both the conversation context and the current turn. We show that LSTM networks with sentence-level attention on context and current turn, as well as the conditional LSTM network, outperform the LSTM model that reads only the current turn. As conversation context, we consider the prior turn, the succeeding turn, or both. Our computational models are tested on two types of social media platforms: Twitter and discussion forums. We discuss several differences between these data sets, ranging from their size to the nature of the gold-label annotations. To address the latter two issues, we present a qualitative analysis of the attention weights produced by the LSTM models (with attention) and discuss the results compared with human performance on the two tasks.

Opportunities for Qualitative Analysis of Social Media Platforms in Dietetic Research and Practice

2020 ◽  
pp. 1-5
Author(s):  
Meghan Lynch ◽  
Irena Knezevic ◽  
Kennedy Laborde Ryan
Keyword(s):  
Social Media ◽  
Low Cost ◽  
Online News ◽  
News Article ◽  
Discussion Forums ◽  
Research And Practice ◽  
The Public ◽  
Related Data ◽  
Social Media Platforms ◽  
Article Comment

To date, most qualitative knowledge about individual eating patterns and the food environment has been derived from traditional data collection methods, such as interviews, focus groups, and observations. However, there currently exists a large source of nutrition-related data in social media discussions that have the potential to provide opportunities to improve dietetic research and practice. Qualitative social media discussion analysis offers a new tool for dietetic researchers and practitioners to gather insights into how the public discusses various nutrition-related topics. We first consider how social media discussion data come with significant advantages including low-cost access to timely ways to gather insights from the public, while also cautioning that social media data have limitations (e.g., difficulty verifying demographic information). We then outline 3 types of social media discussion platforms in particular: (i) online news article comment sections, (ii) food and nutrition blogs, and (iii) discussion forums. We discuss how each different type of social media offers unique insights and provide a specific example from our own research using each platform. We contend that social media discussions can contribute positively to dietetic research and practice.

scholarly journals CASM: A Deep-Learning Approach for Identifying Collective Action Events with Text and Image Data from Social Media

2019 ◽  
Vol 49 (1) ◽  
pp. 1-57 ◽  
Author(s):  
Han Zhang ◽  
Jennifer Pan
Keyword(s):  
Neural Networks ◽  
Social Media ◽  
Collective Action ◽  
Short Term Memory ◽  
Image Data ◽  
Data Sets ◽  
Event Analysis ◽  
Text Data ◽  
Out Of Sample ◽  
Media Reports

Protest event analysis is an important method for the study of collective action and social movements and typically draws on traditional media reports as the data source. We introduce collective action from social media (CASM)—a system that uses convolutional neural networks on image data and recurrent neural networks with long short-term memory on text data in a two-stage classifier to identify social media posts about offline collective action. We implement CASM on Chinese social media data and identify more than 100,000 collective action events from 2010 to 2017 (CASM-China). We evaluate the performance of CASM through cross-validation, out-of-sample validation, and comparisons with other protest data sets. We assess the effect of online censorship and find it does not substantially limit our identification of events. Compared to other protest data sets, CASM-China identifies relatively more rural, land-related protests and relatively few collective action events related to ethnic and religious conflict.

LSTM-based soft sensor design for oxygen content of flue gas in coal-fired power plant

2020 ◽  
pp. 014233122093239
Author(s):  
Hongguang Pan ◽  
Tao Su ◽  
Xiangdong Huang ◽  
Zheng Wang
Keyword(s):  
Power Plant ◽  
Oxygen Content ◽  
Flue Gas ◽  
Oxygen Sensor ◽  
Short Term Memory ◽  
Support Vector ◽  
Data Sets ◽  
Auxiliary Variables ◽  
Data Set ◽  
Lstm Network

To address problems of high cost, complicated process and low accuracy of oxygen content measurement in flue gas of coal-fired power plant, a method based on long short-term memory (LSTM) network is proposed in this paper to replace oxygen sensor to estimate oxygen content in flue gas of boilers. Specifically, first, the LSTM model was built with the Keras deep learning framework, and the accuracy of the model was further improved by selecting appropriate super-parameters through experiments. Secondly, the flue gas oxygen content, as the leading variable, was combined with the mechanism and boiler process primary auxiliary variables. Based on the actual production data collected from a coal-fired power plant in Yulin, China, the data sets were preprocessed. Moreover, a selection model of auxiliary variables based on grey relational analysis is proposed to construct a new data set and divide the training set and testing set. Finally, this model is compared with the traditional soft-sensing modelling methods (i.e. the methods based on support vector machine and BP neural network). The RMSE of LSTM model is 4.51% lower than that of GA-SVM model and 3.55% lower than that of PSO-BP model. The conclusion shows that the oxygen content model based on LSTM has better generalization and has certain industrial value.

scholarly journals Machine Learning in Detecting COVID-19 Misinformation on Twitter

2021 ◽  
Vol 13 (10) ◽  
pp. 244
Author(s):  
Mohammed N. Alenezi ◽  
Zainab M. Alqenaei
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Short Term Memory ◽  
K Nearest Neighbors ◽  
Daily Lives ◽  
Frontline Workers ◽  
Detection Model ◽  
Social Media Platforms ◽  
Long Short Term Memory ◽  
Types Of Information

Social media platforms such as Facebook, Instagram, and Twitter are an inevitable part of our daily lives. These social media platforms are effective tools for disseminating news, photos, and other types of information. In addition to the positives of the convenience of these platforms, they are often used for propagating malicious data or information. This misinformation may misguide users and even have dangerous impact on society’s culture, economics, and healthcare. The propagation of this enormous amount of misinformation is difficult to counter. Hence, the spread of misinformation related to the COVID-19 pandemic, and its treatment and vaccination may lead to severe challenges for each country’s frontline workers. Therefore, it is essential to build an effective machine-learning (ML) misinformation-detection model for identifying the misinformation regarding COVID-19. In this paper, we propose three effective misinformation detection models. The proposed models are long short-term memory (LSTM) networks, which is a special type of RNN; a multichannel convolutional neural network (MC-CNN); and k-nearest neighbors (KNN). Simulations were conducted to evaluate the performance of the proposed models in terms of various evaluation metrics. The proposed models obtained superior results to those from the literature.

scholarly journals A Framework to process Text Data of Web Discussion Forums A Study of LisLinks

2019 ◽  
Vol 39 (06) ◽  
pp. 315-321
Author(s):  
Mohit Garg ◽  
Uma Kanjilal
Keyword(s):  
Social Media ◽  
Unstructured Data ◽  
The Internet ◽  
Discussion Forums ◽  
Text Data ◽  
Knowledge Repository ◽  
Social Media Platforms ◽  
Statistical Programming ◽  
Application Programming ◽  
Programming Interfaces

Nowadays, people use the internet for both seeking and disseminating information in a collaborative way on various social media platforms like Quora, Yahoo Answers, LisLinks Forum, etc. This social interaction on different topics makes these platforms as a knowledge repository. Evaluation of these repositories can help to understand various trends. However, this evaluation is a challenging task because of unstructured data and the unavailability of application programming interfaces for the harvesting of a dataset. This study presented a framework to harvest and pre-processing of data available on LisLinks Forum. The proposed framework is implemented using statistical programming language R. The fourteen metadata elements were defined for the discussion forums. The framework automatically harvest and pre-process relevant data of posts.

scholarly journals HPCWMF: A Hybrid Predictive Cloud Workload Management Framework Using Improved LSTM Neural Network

2020 ◽  
Vol 20 (4) ◽  
pp. 55-73
Author(s):  
K. Dinesh Kumar ◽  
E. Umamaheswari
Keyword(s):  
Short Term Memory ◽  
Hybrid Approach ◽  
Differential Evolution Algorithm ◽  
Memory Loss ◽  
Data Sets ◽  
Workload Prediction ◽  
Workload Management ◽  
Management Framework ◽  
Multilayer Perceptron Network ◽  
Lstm Network

AbstractFor cloud providers, workload prediction is a challenging task due to irregular incoming workloads from users. Accurate workload prediction is essential for scheduling the resources to the cloud applications. Thus, in this paper, the authors propose a predictive cloud workload management framework to estimate the needed resources in advance based on a hybrid approach, which is a combination of an improved Long Short-Term Memory (LSTM) network and a multilayer perceptron network. By improving the traditional LSTM architecture by using opposition-based differential evolution algorithm and dropout technique on recurrent connection without memory loss, the proposed approach has the ability to perform a better prediction process. A novel hybrid predictive approach is aiming at enhancing the prediction performance of the cloud workload. Finally, the authors measure the proposed approach’s effectiveness under benchmark data sets of NASA and Saskatchewan servers. The experimental results proved that the proposed approach outperforms the other conventional methods.

scholarly journals Sosiaalisen median kommenttien analyysi multimodaalisesta näkökulmasta digitaalisen journalismin ja poliittisen viestinnän tutkimuksessa

2021 ◽  
Vol 40 (3) ◽  
Author(s):  
Margareta Salonen ◽  
Elisa Kannasto ◽  
Laura Paatelainen
Keyword(s):  
Social Media ◽  
Content Analysis ◽  
Literature Review ◽  
Political Communication ◽  
Qualitative Content Analysis ◽  
Data Sets ◽  
Individual Mode ◽  
Digital Journalism ◽  
Social Media Platforms ◽  
The One

Societal discussions flow on social media platforms that are studied by researchers in multiple ways and through various kinds of data sets that are extracted from them. In the studies of these discussions, multimodality unravels the semiotic modes that are communication resources through which meanings are socially and culturally created and expressed. In addition, the viewpoint of affordances can be used for viewing the functions of social media platforms and their discussions. Furthermore, this review was conducted to better understand how social media comments are researched from the perspective of multimodality in the context of digital journalism and political communication. A systematic literature review and qualitative content analysis were used as methods. The review discovered that the studies under review were not that high in multimodality and that text as an individual mode was the most common one. Furthermore, Twitter was the most researched platform and the one where the use of modes was more thoroughly explained.

scholarly journals Social Media based Simulation Models for Understanding Disease Dynamics

2018 ◽  
Author(s):  
Ting Hua ◽  
Chandan K Reddy ◽  
Lei Zhang ◽  
Lijing Wang ◽  
Liang Zhao ◽  
...  
Keyword(s):  
Social Media ◽  
Real Time ◽  
Computational Models ◽  
Simulation Models ◽  
Population Level ◽  
Contact Network ◽  
Real World Data ◽  
Computational Epidemiology ◽  
Individual Health Status ◽  
Social Media Platforms

In this modern era, infectious diseases, such as H1N1, SARS, and Ebola, are spreading much faster than any time in history. Efficient approaches are therefore desired to monitor and track the diffusion of these deadly epidemics. Traditional computational epidemiology models are able to capture the disease spreading trends through contact network, however, one unable to provide timely updates via real-world data. In contrast, techniques focusing on emerging social media platforms can collect and monitor real-time disease data, but do not provide an understanding of the underlying dynamics of ailment propagation. To achieve efficient and accurate real-time disease prediction, the framework proposed in this paper combines the strength of social media mining and computational epidemiology. Specifically, individual health status is first learned from user's online posts through Bayesian inference, disease parameters are then extracted for the computational models at population-level, and the outputs of computational epidemiology model are inversely fed into social media data based models for further performance improvement. In various experiments, our proposed model outperforms current disease forecasting approaches with better accuracy and more stability.

scholarly journals The use of social media among health professionals to facilitate chronic disease self-management with their patients: A systematic review

2018 ◽  
Vol 4 ◽  
pp. 205520761877141 ◽  
Author(s):  
Gino De Angelis ◽  
George A Wells ◽  
Barbara Davies ◽  
Judy King ◽  
Shirin M Shallwani ◽  
...  
Keyword(s):  
Social Media ◽  
Chronic Disease ◽  
Health Professionals ◽  
Ease Of Use ◽  
Self Management ◽  
Discussion Forums ◽  
Perceived Ease Of Use ◽  
Collaborative Project ◽  
Social Media Platforms ◽  
Use Of Social Media

Objective The objective of this systematic review was to summarize the evidence pertaining to the use of social media by health professionals to facilitate chronic disease self-management with their patients. Methods A systematic approach was used to retrieve and extract relevant data. A total of 5163 citations were identified, of which seven unique studies met criteria for inclusion; one was a randomized controlled trial, two were prospective cohort studies, and four were qualitative studies. The following social media platforms were evaluated: discussion forums (6 studies) and collaborative project (1 study). Results The available evidence suggests that health professionals perceived discussion forums and collaborative projects to be useful social media platforms to facilitate chronic disease self-management with patients. No relevant evidence was found regarding the use of other social media platforms. Most studies indicated positive findings regarding health professionals’ intention to use discussion forums, while the one study that used a collaborative project also indicated positive findings with its perceived ease of use as health professionals felt that it was useful to facilitate chronic disease self-management with patients. Mixed findings were seen in regards to health professionals’ perceived ease of use of discussion forums. The most common barrier to using social media platforms was the lack of time in health professionals’ schedules. Conclusions Discussion forums and collaborative projects appear to be promising resources for health professionals to assist their patients in self-managing their chronic conditions; however, further research comparing various social media platforms is needed.

scholarly journals Perceptions and Sentiments About Electronic Cigarettes on Social Media Platforms: Systematic Review (Preprint)

2019 ◽  
Author(s):  
Misol Kwon ◽  
Eunhee Park
Keyword(s):  
Social Media ◽  
Online Discussion ◽  
Electronic Cigarettes ◽  
Future Research ◽  
Discussion Forums ◽  
Online Forums ◽  
Online Discussion Forums ◽  
Social Media Platforms ◽  
Potential Risks ◽  
People’S Perceptions

BACKGROUND Electronic cigarettes (e-cigarettes) have been widely promoted on the internet, and subsequently, social media has been used as an important informative platform by e-cigarette users. Beliefs and knowledge expressed on social media platforms have largely influenced e-cigarette uptake, the decision to switch from conventional smoking to e-cigarette smoking, and positive and negative connotations associated with e-cigarettes. Despite this, there is a gap in our knowledge of people’s perceptions and sentiments on e-cigarettes as depicted on social media platforms. OBJECTIVE This study aimed to (1) provide an overview of studies examining the perceptions and sentiments associated with e-cigarettes on social media platforms and online discussion forums, (2) explore people’s perceptions of e-cigarette therein, and (3) examine the methodological limitations and gaps of the included studies. METHODS Searches in major electronic databases, including PubMed, Cumulative Index of Nursing and Allied Health Literature, EMBASE, Web of Science, and Communication and Mass Media Complete, were conducted using the following search terms: “electronic cigarette,” “electronic vaporizer,” “electronic nicotine,” and “electronic nicotine delivery systems” combined with “internet,” “social media,” and “internet use.” The studies were selected if they examined participants’ perceptions and sentiments of e-cigarettes on online forums or social media platforms during the 2007-2017 period. RESULTS A total of 21 articles were included. A total of 20 different social media platforms and online discussion forums were identified. A real-time snapshot and characteristics of sentiments, personal experience, and perceptions toward e-cigarettes on social media platforms and online forums were identified. Common topics regarding e-cigarettes included positive and negative health effects, testimony by current users, potential risks, benefits, regulations associated with e-cigarettes, and attitude toward them as smoking cessation aids. CONCLUSIONS Although perceptions among social media users were mixed, there were more positive sentiments expressed than negative ones. This study particularly adds to our understanding of current trends in the popularity of and attitude toward e-cigarettes among social media users. In addition, this study identified conflicting perceptions about e-cigarettes among social media users. This suggests that accurate and up-to-date information on the benefits and risks of e-cigarettes needs to be disseminated to current and potential e-cigarette users via social media platforms, which can serve as important educational channels. Future research can explore the efficacy of social media–based interventions that deliver appropriate information (eg, general facts, benefits, and risks) about e-cigarettes. CLINICALTRIAL PROSPERO CRD42019121611; https://tinyurl.com/yfr27uxs

Sign in / Sign up

Export Citation Format

Share Document