Detect Extreme Sentiments on Social Networks using BERT

Online social networking platforms allow people to freely express their ideas, opinions, and emotions negatively or positively. Previous studies have examined user’s sentiments on these platforms to study their behaviour in different contexts and purposes. The mechanism of collecting public opinion information has attracted researchers to automatically classify the polarity of public opinions based on the use of concise language in messages, such as tweets, by analyzing social media data. In this paper, we extend the preceding work [1], by proposing an unsupervised approach to automatically detect extreme opinions/posts in social networks. We have evaluated our performance on five different social network and media datasets. In this work, we use the semi-supervised approach BERT to check the accuracy of our classified dataset. The latter task shows that, in these datasets, posts that were previously classified as negative or positive are, in fact, extremely negative or positive in many cases.

ANALYSIS OF VARIOUS APPROACHES OF ABUSIVE TEXT DETECTION IN ONLINE SOCIAL NETWORKS

Social media is one of the most influential tool for sharing information across different regions among different users .The people sharing their interests in various aspects in online social networking platforms like Facebook, twitter etc. Therefore the usage of hate text steadily increasing. Nowadays it has been reviled unfair behavior of the users in social networking sites. The existence of abusive text on different online social networking platforms and identification of such text is a big challenging task. To understand the complexity of language constructs in different languages is very difficult .Already lot of research work has completed in English language. This paper gives detail analysis of detecting hate text in various languages Hindi, urdu, Arabic, Bengali, Telugu. We incorporated various kinds of ML and DL based algorithms to identify hate text in OSN’s. A review is done related to different classifiers where a comparison made between different models of ML, DL algorithms. Finally finds the accurate method to classify the text is offensive or not by finding the parameters i.e. accuracy and F1score

Quality Expectations of Social Networking Sites: A Comparative Study Between New Zealand and Vietnam

<p>Online Social Networking (OSN) websites have been growing fast and their success is decided by customers’ satisfaction. User satisfaction can be measured using a number of popular frameworks such as SERVQUAL, SERVPERF, SiteQual, and WebQual. These frameworks are often used as guidelines when designing, implementing, and assessing quality of websites in general and can also be used to measure the quality of online social networking websites. Besides, there are additional factors that should be taken into consideration when assessing user satisfaction such as demographic differences and cultural differences. The present research project aims to investigate and test the linkage between culture and user satisfaction on the online social networking websites. This is achieved by conducting an analysis on the basis of a survey in two different countries – New Zealand and Vietnam. The research project’s principal component analysis follows Hofstede’s six cultural dimensions and the modified framework of assessing online social networking sites quality conducted by Rizavi, Ali, and Rizavi in 2011. The results suggest that users’ quality expectation of Social Networking Sites (SNS) in New Zealand and in Vietnam may be influenced by cultural differences.</p>

Quality Expectations of Social Networking Sites: A Comparative Study Between New Zealand and Vietnam

<p>Online Social Networking (OSN) websites have been growing fast and their success is decided by customers’ satisfaction. User satisfaction can be measured using a number of popular frameworks such as SERVQUAL, SERVPERF, SiteQual, and WebQual. These frameworks are often used as guidelines when designing, implementing, and assessing quality of websites in general and can also be used to measure the quality of online social networking websites. Besides, there are additional factors that should be taken into consideration when assessing user satisfaction such as demographic differences and cultural differences. The present research project aims to investigate and test the linkage between culture and user satisfaction on the online social networking websites. This is achieved by conducting an analysis on the basis of a survey in two different countries – New Zealand and Vietnam. The research project’s principal component analysis follows Hofstede’s six cultural dimensions and the modified framework of assessing online social networking sites quality conducted by Rizavi, Ali, and Rizavi in 2011. The results suggest that users’ quality expectation of Social Networking Sites (SNS) in New Zealand and in Vietnam may be influenced by cultural differences.</p>

Facebook me!: Initial insights in favour of using social networking as a tool for translator training

This paper argues in favour of integrating and using online social networking, more specifically Facebook, within the translation classroom. This has numerous benefits in terms of aptly preparing trainees for the marketplace and also helping to foster a classroom community by encouraging a collaborative learning environment. A descriptive analysis of five undergraduate courses suggests that using online social networking as a teaching strategy has a significant impact: from engaging students through collaborative translation projects, to peer-reviewing assignments, to establishing ‘telepresence’, Facebook allows the trainer to ‘connect’ the classroom.

Roles of Psychosocial Factors on the Association Between Online Social Networking Use Intensity and Depressive Symptoms Among Adolescents: Prospective Cohort Study

Background The potential mechanisms underlying the association between online social networking use intensity and depressive symptoms are unclear and underresearched. Objective We aimed to investigate the potential roles of interpersonal psychosocial factors on the association between online social networking use intensity and depressive symptoms among early adolescents. Methods A total of 4237 adolescents from a 9-month longitudinal study were included. Score changes (indicated as △) for the social function use intensity (SFUI) and entertainment function use intensity (EFUI) subscales of the Online Social Networking Activity Intensity Scale and for friendship quality, perceived family support, perceived friend support, parent–adolescent conflict, social nonconfidence, and depressive symptoms were analyzed. The potential mediation effects of unfavorable psychosocial factors and suppression effects of favorable psychosocial factors on the association of △SFUI with △CES-D and the association of △EFUI with △CES-D were tested using hierarchical regression models. Results The association between △SFUI and △CES-D was partially mediated by △mother–adolescent conflict (mediation effect size 5.11%, P=.02) and △social nonconfidence (mediation effect size 20.97%, P<.001) but partially suppressed by △friendship quality, △perceived family support, and △perceived friend support, with suppression effects of –0.011 (P=.003), –0.009 (P=.003), and –0.022 (P<.001), respectively. The association between △EFUI and △CES-D was partially mediated by △social nonconfidence (mediation effect size 30.65%, P<.001) but partially suppressed by △perceived family support and △perceived friend support, with suppression effects of –0.036 (P<.001) and –0.039 (P<.001), respectively. Conclusions The association between online social networking use intensity and depressive symptoms was partially mediated through the indirect increase in social nonconfidence and mother–adolescent conflict; however, better perceived social support and friendship quality would partially compensate for the harmful impact of online social networking use intensity on depressive symptoms among early adolescents.

RELATIONSHIP BETWEEN ONLINE SOCIAL NETWORKING SITES & APPLICATIONS AND UNIVERSITY STUDENTS’ ACHIEVEMENT

Purpose of the study: This study analysed the relationship between Online Social Networking websites and Applications (OSNSA) and university students' academic achievement. Methodology: In the present research, a quantitative research design was followed, and a survey research type of descriptive research was used in this study. Data were collected from 150 university students using a modified version of The Online Social Networking Sites scale (Paul, Baker, & Cochran, 2012). The correlation was calculated by using the SPSS 22nd version. Main Findings: The findings revealed a connection between students' time spent on OSNSA and their academic outcomes. The time spent on OSNSA was observed to be strongly influenced by the attention span of the students. In particular, we find that a lower academic output is provided by the greater time spent on OSNSA. Also, attention span was strongly associated with features that influence students' academic success, such as their perceptions of society's social networking outlook, their advantages, and disadvantages of OSNSA, use of OSNSA, etc. Applications of this study: Many students spend most of their hours on online networking pages, like Facebook, Tok-Tok, and WhatsApp, which are the most common, considering the harmful impact of social media. In certain eyes, social media has embedded a detrimental effect, but it is also used to acquire vital knowledge and social skills for pupils. Novelty/Originality of this study: This research is different from other studies because it explores the relationship between social networking sites & the application and academic performance of university students belonging from South Punjab by exploring various facets of OSNSA which ultimately promote university students’ academic performance.