Artificial Intelligence for Online Safety

Dear Colleagues,

Nearly every aspect of our lives is impacted by the Internet, including entertainment, education, health, commerce, government, social interaction, and more. The profound impact that these recent changes are having on our lives requires us to rethink how we make decisions in these areas. In addition to questions about costs and benefits, users also face important questions about trust and security. For example, we often find cases where people avoid online banking or buying products online because they fear becoming victims of fraudulent activity. In addition, informed users are increasingly reluctant to trust information sources. Therefore, it is of utmost importance to create a trustworthy environment for users—an environment where both users and contents are trustworthy.

When fighting threats, companies face a dilemma, given that no system is perfect. In e-commerce, on the one hand, threats and the derived losses should be reduced; on the other hand, users neither want to be accused of fraud nor treated like criminals. In other areas, such as e-health, the problems associated with data abuse and security leaks could even result in more severe damages than purely financial matters. Finally, platforms that allow users to share information and non-curated content have recently faced the complex trade-off between free expression and moderation. In this latter case, the spread of misinformation poses a threat to society, health, and even democracy.           

Given the exponential growth and exploitation of these vulnerabilities for businesses and societies, online service providers are looking for automated solutions that can mitigate problems like automated threats detection, malicious user activity detection, automated content curation.

In recent years, AI has taken on an increasingly central role in threats prevention. The reason why AI techniques are so popular and widely used in the threats and fraud detection industry is due to (i) their fast computational power in analyzing and processing data and extracting new patterns; (ii) their scalability, as models become more accurate and effective in prediction the more data they receive; (iii) their efficiency in obtaining results compared to manual efforts.

Although both industry and academia fields have always invested significant efforts in tackling the above-mentioned problems, we have identified a gap between the fields. On the one hand, industry, which is affected firsthand by these problems, has a wealth of data and suboptimal solutions (which are not always shared) to mitigate these risks. On the other hand, academia has cutting-edge solutions, but often has limited access to data and users.

This Special Issue aims at bringing together research from a wide array of disciplines (mathematics, computer science, economy, philosophy, social science) to (i) understand the cases and motivations of fraudulent activities in online environments, (ii) find AI solutions to detect and analyze threats, malicious activities and the spread and misinformation, and (iii) derive means to prevent them.

We invite the submission of ongoing and completed research work with a particular focus on the following topics:

• User Modeling of Fraudulent and Malicious Users
• Features Engineering in the Online Threats Detection Domain
• Outlier and Anomaly Detection
• Fraud Detection
• Detecting, preventing and predicting anomalies in Web data (e.g., fake content, spam, algorithmic and data biases)
• Fraud Detection in the Streaming Domain
• Distributed Fraud Detection Systems
• Malicious user activity in Web-based systems
• Safeguarding and governance of the Web, including anonymity, security and trust
• Accountable use of the Web
• Online Safety in the Medical Domain

Dr. Marco Fisichella
Guest Editor

Dr. Antonia Russo
Guest Editor Assistant

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Big Data and Cognitive Computing is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• fraud detection
• outlier and anomaly detection
• anonymity
• security and trust
• privacy
• accountability and auditability
• federated learning