Cross-Domain Spam Detection in Social Media: A Survey

2019 ◽  
pp. 98-112
Author(s):  
Deepali Dhaka ◽  
Monica Mehrotra
Keyword(s):  
Social Media ◽  
Spam Detection ◽  
Cross Domain

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 Spam, a Digital Pollution and Ways to Eradicate It

2019 ◽  
Vol 9 (2) ◽  
pp. 2630-2638
Keyword(s):  
Social Media ◽  
Spam Detection ◽  
Combination Of Methods ◽  
Survey Analyses

Due to the growing popularity of the microblogging and networking sites like twitter, Gmail, Facebook etc., there has been an increase in the number of spammers. Spammers on Twitter seem to be more dangerous than the mail spammers as they exploit the limitation on the characters of Twitter for their own purposes. Spammers have also become creative in framing their content to cleverly escape the classifiers. This survey is thus mainly used to discuss and analyze the recent research that had been put forth regarding the spam detection in social media sites such as Twitter. This survey analyses the papers that tackled various problems faced on Twitter and the problems faced by the methods that have already been presented before. We then compared all the methods present in the papers to see which method or combination of methods could give the best result in detecting spam.

Adversarial Cross-domain Community Question Retrieval

2022 ◽  
Vol 21 (3) ◽  
pp. 1-22
Author(s):  
Aibo Guo ◽  
Xinyi Li ◽  
Ning Pang ◽  
Xiang Zhao
Keyword(s):  
Social Media ◽  
Information Sharing ◽  
Information Access ◽  
Professional Knowledge ◽  
Feature Space ◽  
Online Information ◽  
Cross Domain ◽  
Adversarial Training ◽  
Sentence Matching ◽  
Target Data

Community Q&A forum is a special type of social media that provides a platform to raise questions and to answer them (both by forum participants), to facilitate online information sharing. Currently, community Q&A forums in professional domains have attracted a large number of users by offering professional knowledge. To support information access and save users’ efforts of raising new questions, they usually come with a question retrieval function, which retrieves similar existing questions (and their answers) to a user’s query. However, it can be difficult for community Q&A forums to cover all domains, especially those emerging lately with little labeled data but great discrepancy from existing domains. We refer to this scenario as cross-domain question retrieval. To handle the unique challenges of cross-domain question retrieval, we design a model based on adversarial training, namely, X-QR , which consists of two modules—a domain discriminator and a sentence matcher. The domain discriminator aims at aligning the source and target data distributions and unifying the feature space by domain-adversarial training. With the assistance of the domain discriminator, the sentence matcher is able to learn domain-consistent knowledge for the final matching prediction. To the best of our knowledge, this work is among the first to investigate the domain adaption problem of sentence matching for community Q&A forums question retrieval. The experiment results suggest that the proposed X-QR model offers better performance than conventional sentence matching methods in accomplishing cross-domain community Q&A tasks.

scholarly journals A STUDY OF SPAM DETECTION ALGORITHM ON SOCIAL MEDIA NETWORKS

2014 ◽  
Vol 10 (10) ◽  
pp. 2135-2140 ◽  
Author(s):  
Saini Jacob Soman ◽  
S. Murugappan
Keyword(s):  
Social Media ◽  
Detection Algorithm ◽  
Spam Detection ◽  
Social Media Networks

Sentiment search: an emerging trend on social media monitoring systems

2014 ◽  
Vol 66 (5) ◽  
pp. 553-580 ◽  
Author(s):  
Tung Thanh Nguyen ◽  
Tho Thanh Quan ◽  
Tuoi Thi Phan
Keyword(s):  
Social Media ◽  
Sentiment Analysis ◽  
Real Data ◽  
Content Type ◽  
Social Monitoring ◽  
Cross Domain ◽  
Linguistic Rules ◽  
Domain Ontologies ◽  
Media Monitoring ◽  
Media Channels

Purpose – The purpose of this paper is to discuss sentiment search, which not only retrieves data related to submitted keywords but also identifies sentiment opinion implied in the retrieved data and the subject targeted by this opinion. Design/methodology/approach – The authors propose a retrieval framework known as Cross-Domain Sentiment Search (CSS), which combines the usage of domain ontologies with specific linguistic rules to handle sentiment terms in textual data. The CSS framework also supports incrementally enriching domain ontologies when applied in new domains. Findings – The authors found that domain ontologies are extremely helpful when CSS is applied in specific domains. In the meantime, the embedded linguistic rules make CSS achieve better performance as compared to data mining techniques. Research limitations/implications – The approach has been initially applied in a real social monitoring system of a professional IT company. Thus, it is proved to be able to handle real data acquired from social media channels such as electronic newspapers or social networks. Originality/value – The authors have placed aspect-based sentiment analysis in the context of semantic search and introduced the CSS framework for the whole sentiment search process. The formal definitions of Sentiment Ontology and aspect-based sentiment analysis are also presented. This distinguishes the work from other related works.

scholarly journals Spam Detection Framework using ML Algorithm

2020 ◽  
Vol 8 (6) ◽  
pp. 5326-5329
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Social Networking ◽  
Information Sharing ◽  
Detection System ◽  
Reliable Data ◽  
Machine Learning Algorithms ◽  
Spam Detection ◽  
Function Type ◽  
Use Of Social Media

The current use of social media has created incomparable amounts of social data, as it is a cheap and popular information sharing communication platform. Nowadays, a huge percentage of people depend on the accessible material on social networking in their choices (e.g. comments and suggestions about a subject or product). This feature on exchanging knowledge with a wide number of users has quickly prompted social spammers to exploit the network of confidence to distribute spam messages and support personal forums, advertising, phishing, scams and so on. Identifying these spammers and spam material is a hot subject of study, and while large amounts of experiments have recently been conducted to this end, so far the methodologies are only barely able to identify spam feedback, and none of them demonstrates the value of each derived function type. In this study, we have suggested a machine learning-based spam detection system that determines whether or not a specific message in the dataset is spam using a set of machine learning algorithms. Four main features have been used; including user-behavioral, user-linguistic, reviewbehavioral and review-linguistic, to improve the spam detection process and to gather reliable data

scholarly journals Spam Detection Framework for Twitter using ML

2020 ◽  
Vol 9 (6) ◽  
pp. 216-219
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Bayes Theorem ◽  
Spam Detection ◽  
Bayes Classifier ◽  
Common Mode ◽  
Basic Framework ◽  
Trending Topics ◽  
Twitter Users ◽  
Promotional Messages

Spam has become one of the growing issues in social media websites. Some of the users in these websites creates spam news. Coming to twitter, Users inject tweets in trending topics and replies with promotional messages providing links. A large amount of spam has been noticied in twitter. It is necessary to identify these spams tweets in a twitter stream. Now a days ,a big part of people rely on content available in social media in their decisions, so detecting and deleting these spam details is very important. A basic framework is suggested to detect malicious account holders in twitter..At present to detect these spam users or accounts there are methods which are based on content based features, Graph based features. The system which is going to be created works on machine learning based algorithms. These algorithms help to give accurate results. In this system algorithm named Naïve Bayes classifier algorithm is going to be used. This algorithm is said to be combination of many other principles relyingupon “Bayes theorem” wherein the methods share a common mode of working.

Sign in / Sign up

Export Citation Format

Share Document