scholarly journals Joint Topical Word Embedding for Detecting Drift in Social Media Text

2020 ◽  
Author(s):  
VIJAYARANI J ◽  
Geetha T.V.
Keyword(s):  
Social Media ◽  
Topic Model ◽  
Rate Of Change ◽  
Word Embedding ◽  
Human Interaction ◽  
Langevin Dynamic ◽  
Context Vector ◽  
The Social ◽  
Topic Distribution ◽  
Social Media Text

Abstract Social media texts like tweets and blogs are collaboratively created by human interaction. Fast change in trends leads to topic drift in the social media text. This drift is usually associated with words and hashtags. However, geotags play an important part in determining topic distribution with location context. Rate of change in the distribution of words, hashtags and geotags cannot be considered as uniform and must be handled accordingly. This paper builds a topic model that associates topic with a mixture of distributions of words, hashtags and geotags. Stochastic gradient Langevin dynamic model with varying mini-batch sizes is used to capture the changes due to the asynchronous distribution of words and tags. Topical word embedding with co-occurrence and location contexts are specified as hashtag context vector and geotag context vector respectively. These two vectors are jointly learned to yield topical word embedding vectors related to tags context. Topical word embeddings over time conditioned on hashtags and geotags predict, location-based topical variations effectively. When evaluated with Chennai and UK geolocated Twitter data, the proposed joint topical word embedding model enhanced by the social tags context, outperforms other methods.

scholarly journals Joint Topical Word Embedding for Detecting Drift in Social Media Text

2020 ◽  
Author(s):  
VIJAYARANI J ◽  
Geetha T.V.
Keyword(s):  
Social Media ◽  
Topic Model ◽  
Rate Of Change ◽  
Word Embedding ◽  
Human Interaction ◽  
Langevin Dynamic ◽  
Context Vector ◽  
The Social ◽  
Topic Distribution ◽  
Social Media Text

Abstract Social media texts like tweets and blogs are collaboratively created by human interaction. Fast change in trends leads to topic drift in the social media text. This drift is usually associated with words and hashtags. However, geotags play an important part in determining topic distribution with location context. Rate of change in the distribution of words, hashtags and geotags cannot be considered as uniform and must be handled accordingly. This paper builds a topic model that associates topic with a mixture of distributions of words, hashtags and geotags. Stochastic gradient Langevin dynamic model with varying mini-batch sizes is used to capture the changes due to the asynchronous distribution of words and tags. Topical word embedding with co-occurrence and location contexts are specified as hashtag context vector and geotag context vector respectively. These two vectors are jointly learned to yield topical word embedding vectors related to tags context. Topical word embeddings over time conditioned on hashtags and geotags predict, location-based topical variations effectively. When evaluated with Chennai and UK geolocated Twitter data, the proposed joint topical word embedding model enhanced by the social tags context, outperforms other methods.

scholarly journals Joint Topical Word Embedding for Detecting Drift in Social Media Text

2020 ◽  
Author(s):  
VIJAYARANI J ◽  
Geetha T.V.
Keyword(s):  
Social Media ◽  
Topic Model ◽  
Rate Of Change ◽  
Word Embedding ◽  
Human Interaction ◽  
Langevin Dynamic ◽  
Context Vector ◽  
The Social ◽  
Topic Distribution ◽  
Social Media Text

Abstract Social media texts like tweets and blogs are collaboratively created by human interaction. Fast change in trends leads to topic drift in the social media text. This drift is usually associated with words and hashtags. However, geotags play an important part in determining topic distribution with location context. Rate of change in the distribution of words, hashtags and geotags cannot be considered as uniform and must be handled accordingly. This paper builds a topic model that associates topic with a mixture of distributions of words, hashtags and geotags. Stochastic gradient Langevin dynamic model with varying mini-batch sizes is used to capture the changes due to the asynchronous distribution of words and tags. Topical word embedding with co-occurrence and location contexts are specified as hashtag context vector and geotag context vector respectively. These two vectors are jointly learned to yield topical word embedding vectors related to tags context. Topical word embeddings over time conditioned on hashtags and geotags predict, location-based topical variations effectively. When evaluated with Chennai and UK geolocated Twitter data, the proposed joint topical word embedding model enhanced by the social tags context, outperforms other methods.

Spam Detection on Social Media Using Semantic Convolutional Neural Network

2018 ◽  
Vol 8 (1) ◽  
pp. 12-26 ◽  
Author(s):  
Gauri Jain ◽  
Manisha Sharma ◽  
Basant Agarwal
Keyword(s):  
Neural Network ◽  
Social Media ◽  
Convolutional Neural Network ◽  
State Of The Art ◽  
Spam Detection ◽  
Learning Technology ◽  
The Social ◽  
Social Media Text ◽  
Current Article ◽  
Semantic Layer

This article describes how spam detection in the social media text is becoming increasing important because of the exponential increase in the spam volume over the network. It is challenging, especially in case of text within the limited number of characters. Effective spam detection requires more number of efficient features to be learned. In the current article, the use of a deep learning technology known as a convolutional neural network (CNN) is proposed for spam detection with an added semantic layer on the top of it. The resultant model is known as a semantic convolutional neural network (SCNN). A semantic layer is composed of training the random word vectors with the help of Word2vec to get the semantically enriched word embedding. WordNet and ConceptNet are used to find the word similar to a given word, in case it is missing in the word2vec. The architecture is evaluated on two corpora: SMS Spam dataset (UCI repository) and Twitter dataset (Tweets scrapped from public live tweets). The authors' approach outperforms the-state-of-the-art results with 98.65% accuracy on SMS spam dataset and 94.40% accuracy on Twitter dataset.

scholarly journals News Organizations’ Selective Link Sharing as Gatekeeping : A Structural Topic Model Approach

2019 ◽  
Vol 1 (1) ◽  
pp. 45-78
Author(s):  
Chankyung Pak
Keyword(s):  
Social Media ◽  
Topic Model ◽  
News Stories ◽  
News Websites ◽  
News Organizations ◽  
Computational Data ◽  
Topic Distribution ◽  
Editorial Decisions ◽  
Structural Topic Model ◽  
Model Approach

Abstract To disseminate their stories efficiently via social media, news organizations make decisions that resemble traditional editorial decisions. However, the decisions for social media may deviate from traditional ones because they are often made outside the newsroom and guided by audience metrics. This study focuses on selective link sharing as quasi-gatekeeping on Twitter ‐ conditioning a link sharing decision about news content. It illustrates how selective link sharing resembles and deviates from gatekeeping for the publication of news stories. Using a computational data collection method and a machine learning technique called Structural Topic Model (STM), this study shows that selective link sharing generates a different topic distribution between news websites and Twitter and thus significantly revokes the specialty of news organizations. This finding implies that emergent logic, which governs news organizations’ decisions for social media, can undermine the provision of diverse news.

scholarly journals Exploring Peoples' Perception of Autonomy and Reactance in Everyday AI Interactions

2021 ◽  
Vol 12 ◽  
Author(s):  
Supraja Sankaran ◽  
Chao Zhang ◽  
Henk Aarts ◽  
Panos Markopoulos
Keyword(s):  
Artificial Intelligence ◽  
Decision Making ◽  
Social Media ◽  
Future Development ◽  
Human Interaction ◽  
Potential Threat ◽  
Daily Lives ◽  
The Social ◽  
Human Autonomy ◽  
Design Fiction

Applications using Artificial Intelligence (AI) have become commonplace and embedded in our daily lives. Much of our communication has transitioned from human–human interaction to human–technology or technology-mediated interaction. As technology is handed over control and streamlines choices and decision-making in different contexts, people are increasingly concerned about a potential threat to their autonomy. In this paper, we explore autonomy perception when interacting with AI-based applications in everyday contexts using a design fiction-based survey with 328 participants. We probed if providing users with explanations on “why” an application made certain choices or decisions influenced their perception of autonomy or reactance regarding the interaction with the applications. We also looked at changes in perception when users are aware of AI's presence in an application. In the social media context, we found that people perceived a greater reactance and lower sense of autonomy perhaps owing to the personal and identity-sensitive nature of the application context. Providing explanations on “why” in the navigation context, contributed to enhancing their autonomy perception, and reducing reactance since it influenced the users' subsequent actions based on the recommendation. We discuss our findings and the implications it has for the future development of everyday AI applications that respect human autonomy.

scholarly journals WAJAH “LAMAHOLOT CYBER-SELF”: CATATAN ETNOGRAFIS VIRTUAL ATAS PRESENTASI DIRI ANGGOTA GRUP FACEBOOK “SUARA FLOTIM”

2020 ◽  
Author(s):  
Aurelius RL Teluma ◽  
Rini Kartini
Keyword(s):  
Social Media ◽  
The Self ◽  
Young Generation ◽  
Cultural Interaction ◽  
Ethnographic Methods ◽  
Cultural Economic ◽  
Self Presentation ◽  
The Social ◽  
Social Media Text ◽  
Political Dimensions

Social media is a socio-technological entity. Social media text is a representation of the social, cultural, economic and political dimensions of its users. When social media was known and used by the young generation of the Lamaholot people, there was a cultural interaction with the Lamaholot Lika Telo kinship culture which gave birth to identity and Lamaholot-cyber self. This research is textual-contextual research using virtual ethnographic methods that aim to identify and describe the general characteristics of Lamaholot-cyber social media characteristics. The subjects examined are Lamaholot Facebook users who are members of a Facebook group "Suara Flotim". The virtual ethnographic observations of the self-presentation of Lamaholot Facebookers showed that the cyber Lamaholot was a self and a discursive identity even to be paradoxical entity.

scholarly journals Islam and Social Media: Attitudes and Views

2016 ◽  
Vol 12 (5) ◽  
pp. 221
Author(s):  
Wafa Abu Hatab
Keyword(s):  
Social Media ◽  
English Language ◽  
Human Interaction ◽  
Social Media Platform ◽  
Individual Level ◽  
Positive Attitudes ◽  
The Social ◽  
Media Platform ◽  
And Gender ◽  
Use Of Social Media

<p>Social media has become an integral part of our daily life encapsulating time and place, creating new relations and fostering old ones not only on an individual level but also on social and global ones. This revolution in human interaction was led by the introduction of Facebook in 2004 that was followed by other social media platforms such as Twitter and Instegram. This electronic revolution swept over to reach mobile phones and to introduce new platforms such as WhatsApp and Viber. The present study investigated attitudes and views towards the use of social media in promoting Islam. A random sample of Facebook users was asked to fill in a questionnaire that tackled questions related to their attitudes towards the role of social media in promoting Islam, the linguistic influence of the social media on their English language skills when talking about Islam and the most preferred social media platform. . Respondents were then classified according to education and gender. The study revealed that the social media have affected the way the other is addressed when discussing Islamic topics. Despite some negative stands, the positive attitudes towards social media in promoting Islam prevailed. The views were influenced by the respondents’ age, gender and education. The linguistic influence of the social media on developing English skills was viewed positively. The Facebook was the most preferred social media platform. Further research is recommended on the interrelationships between social factors and views of social media. Code-switching among social media users and the effect on Arabic might be also investigated.</p>

scholarly journals Sentiment analysis of Social Media Text-Emoticon Post with Machine learning Models Contribution Title

2021 ◽  
Vol 2070 (1) ◽  
pp. 012079
Author(s):  
V Jagadishwari ◽  
A Indulekha ◽  
Kiran Raghu ◽  
P Harshini
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Sentiment Analysis ◽  
Online Social Networks ◽  
Data Sets ◽  
Learning Models ◽  
Twitter Data ◽  
The Social ◽  
Social Media Text ◽  
Machine Learning Models

Abstract Social Media is an arena in recent times for people to share their perspectives on a variety of topics. Most of the social interactions are through the Social Media. Though all the Online Social Networks allow users to express their views and opinions in many forms like audio, video, text etc, the most popular form of expression is text, Emoticons and Emojis. The work presented in this paper aims at detecting the sentiments expressed in the Social Media posts. The Machine Learning Models namely Bernoulli Bayes, Multinomial Bayes, Regression and SVM were implemented. All these models were trained and tested with Twitter Data sets. Users on Twitter express their opinions in the form of tweets with limited characters. Tweets also contain Emoticons and Emojis therefore Twitter data sets are best suited for the sentiment analysis. The effect of emoticons present in the tweet is also analyzed. The models are first trained only with the text and then they are trained with text and emoticon in the tweet. The performance of all the four models in both cases are tested and the results are presented in the paper.

scholarly journals An Analysis of the Themes of Refuting Rumors on China's Sina Weibo During the Period of Covid-19

2021 ◽  
Vol 2 (2) ◽  
pp. 174-179
Author(s):  
Difan Guo
Keyword(s):  
Social Media ◽  
Topic Model ◽  
Release Channel ◽  
Sina Weibo ◽  
News Release ◽  
System Disease ◽  
Study Find ◽  
The Social ◽  
Media Platform ◽  
The Government

From the end of 2019 to 2020, there were countless rumors on the Internet related to COVID-19 during the viral epidemic. This study analyzed how government Weibo, the official news release channel of government social media, refuted rumors on China's leading social media platform Sina Weibo during the COVID-19 pandemic outbreak in China. This study used the LDA topic model to model the Weibo text topic and obtain the topics of the rumors that the government Weibo defied. This study find that the five main topics of rumors presented in the anti-rumor Weibo are highly related to the operation of the social system, disease prevention and treatment, and social security.  

Sign in / Sign up

Export Citation Format

Share Document