A survey of event detection techniques in online social networks

Goswami, Anuradha; Kumar, Ajey

doi:10.1007/s13278-016-0414-1

Cited by 58 publications

(50 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One of the main applications of social media mining is the automatic detection of events and behaviors which includes identifying people behavior in real-world events through monitoring their interactions with each other. Researchers can take an advantage of these explosive data to reach substantial insights [21]. This task depends mainly on text mining approaches such as NLP and machine learning algorithms.…”

Section: Automatic Hate Speech Detection In Social Networkmentioning

confidence: 99%

Detection of Hate Speech in Social Networks: A Survey on Multilingual Corpus

Al-Dossari¹

2019

Computer Science &Amp; Information Technology(CS &Amp; IT)

119

View full text Add to dashboard Cite

In social media platforms, hate speech can be a reason of "cyber conflict" which can affect social life in both of individual-level and country-level. Hateful and antagonistic content propagated via social networks has the potential to cause harm and suffering on an individual basis and lead to social tension and disorder beyond cyber space. However, social networks cannot control all the content that users post. For this reason, there is a demand for automatic detection of hate speech. This demand particularly raises when the content is written in complex languages (e.g. Arabic). Arabic text is known with its challenges, complexity and scarcity of its resources. This paper will present a background on hate speech and its related detection approaches. In addition, the recent contributions on hate speech and its related anti-social behaviour topics will be reviewed. Finally, challenges and recommendations for the Arabic hate speech detection problem will be presented.

show abstract

Section: Automatic Hate Speech Detection In Social Networkmentioning

confidence: 99%

Detection of Hate Speech in Social Networks: A Survey on Multilingual Corpus

Al-Dossari¹

2019

Computer Science &Amp; Information Technology(CS &Amp; IT)

119

View full text Add to dashboard Cite

show abstract

“…Conversely, fewer than 10 topics may show small granularity and be insufficient to create a representative analysis (Lugmayr and Grueblbauer 2016). Moreover, most studies on microblogs have predefined the number of topics in their studies, for example, by using 10 or 20 topics (Goswami and Kumar 2016;. For instance, Lo, Chiong, and Cornforth (2016) argued that 20topic LDA models perform better than 10 and 30-topic models.…”

Section: Extracting Topicsmentioning

confidence: 99%

Decoding the sentiment dynamics of online retailing customers: Time series analysis of social media

Ibrahim

Wang

2019

Computers in Human Behavior

View full text Add to dashboard Cite

Decoding the sentiment dynamics of online retailing customers: Time series analysis of social media. Computers in Human Behavior, 96,[32][33][34][35][36][37][38][39][40][41][42][43][44][45]. https://doi. AbstractThe Twittersphere often offers valuable information about current events. However, despite the enormous quantity of tweets regarding online retailing, we know little about customers' perceptions regarding the products and services offered by online retail brands. Therefore, this study focuses on analysing brand-related tweets associated with five leading UK online retailers during the most important sales period of the year, covering Black Friday, Christmas and the New Year's sales events.We explore trends in customer tweets by utilising a combination of data analytics approaches including time series analysis, sentiment analysis and topic modelling to analyse the trends of tweet volume and sentiment and to understand the reasons underlying changes in sentiment. Through the sentiment and time series analyses, we identify several critical time points that lead to significant deviations in sentiment trends. We then use a topic modelling approach to examine the tweets in the period leading up to and following these critical moments to understand what exactly drives these changes in sentiment. The study provides a deeper understanding of online retailing customer behaviour and derives significant managerial insights that are useful for improving online retailing service provision.Ibrahim, Wang, and Bourne 2017). In addition, an overwhelmingly large volume of data (Netzer, Feldman, Goldenberg, and Fresko 2012) and a lack of practical tools with which to analyse unstructured data (Archak, Ghose, and Ipeirotis 2011) also complicate such analysis. Since Twitter content is produced abundantly every day, changes in customer sentiment and the relationship between those changes and emerging issues on Twitter are difficult to capture. The root cause of such changes may lie in a range of factors, including external events, accidents and disasters, and users' daily lives.Therefore, to understand what exactly drives changes in customer sentiment, this study attempts to exploit the content of these real-time data to understand trends in customer tweets. Thus, the research questions are as follows:• What are the volume and sentiment trends of online retailing brand-related tweets on Twitter?• What are the main reasons behind the changes in the volume and sentiment trends of online retail brand-related tweets?• How can online retailers learn from social media users to improve their online retailing service provision?To answer these questions, we analyse changes in customer sentiment and the underlying customer concerns that evolve through the time series. We examine those tweets or comments posted by customers that are directly related to five leading UK retailers namely Amazon, Argos, Asda, John Lewis and Tesco during the period spanning Black Friday, Christmas, Boxing Day and New Year's Day in the United Kingdom. Using a ...

show abstract

“…According to the literature, textual data represents about 80% of the total data on the web [3]. In fact, textual data comes in two styles (e.g., formal and informal) and generated from various sources such as news websites, news feeds, Weblogs, Forums, Emails, Facebook and Twitter [4]. Recently, Facebook has emerged as a powerful source to get knowledge about different real-world events [5].…”

Section: Introductionmentioning

confidence: 99%

“…All together has motivated many researchers from Event Detection (ED) field to introduce several ED models [9]. However, several challenges for building high accuracy ED model for Facebook news posts are identified by different researchers [4]. One important challenge is the high dimensional feature space that contains various kind of features such as redundant, irrelevant and noisy features [10].…”

Section: Introductionmentioning

confidence: 99%

“…These features cause the problem of overfitting for the given ED model, mislead the detection method and eventually decrease the overall accuracy of ED model [11]. The other crucial challenge is designing an unsupervised detection method which is capable to detect and summarize different real-world events from large volume of short text Facebook news posts [4]. Therefore, phases such as pre-processing phase, feature selection phase and event detection phase have become vital phases in building ED model since the model's accuracy is highly influenced by the methods used under those phases [4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Event Detection Model for Facebook News Posts

Al-Dyani,

Ahmad*,

Kamaruddin

2019

IJITEE

View full text Add to dashboard Cite

Event detection has wide application especially in the area of news streams analyzing where there is a need to monitor what events are emerging and affecting people’s lives. This is crucial for public administrations and policy makers to learn from their previous mistakes to make better decisions in the future. Different researchers have introduced several event detection models for Facebook news posts in. However, majority of these models have not provided adequate information about the discovered news events such as location, people and activity. In addition, existing models have ignored the problem of high dimensional feature space which affects the overall detection performance of the models. This research presents a conceptual event detection model for mining events from large volume of short text Facebook news posts and summarize their valuable information. This is crucial for public administrations and policy makers to learn from their previous mistakes to make better decisions in the future. The proposed model includes pre-processing, feature selection, event detection and summarization phases. The pre-processing phase involves several steps to convert unstructured text news posts into structure data. Feature selection phase to select the optimal feature subset. Meanwhile, event detection phase uses these features to construct undirected weighted graph and apply dynamic graph technique to identify the clusters from the graph and then annotate each cluster to its corresponding event. At the end of this paper, several unresolved problems in the construction of event detection model from Facebook news posts are reported to be used as future work for the current study.

show abstract

A survey of event detection techniques in online social networks

Cited by 58 publications

References 72 publications

Detection of Hate Speech in Social Networks: A Survey on Multilingual Corpus

Detection of Hate Speech in Social Networks: A Survey on Multilingual Corpus

Decoding the sentiment dynamics of online retailing customers: Time series analysis of social media

Event Detection Model for Facebook News Posts

Contact Info

Product

Resources

About