Several solutions from the e-Commerce domain can also be utilized in the agriculture sector [28 (link)]. For example, IoT-based applications for safe transportation of sensitive pharmaceutical products in the context of e-Commerce are relevant to similar solutions for the transportation of sensitive agricultural goods [29 ,30 ]. Other popular examples include applications that use ML-based network monitoring tools to detect and classify malicious operations, such as information tampering and DoS attacks. [31 ,32 ]. Network traffic classification solutions for IoT systems can also be adapted to be used in agriculture or other sectors since network traffic attributes have similarities across different domains [33 (link)].
A part of the current research work has also been implemented in the context of the ENSURESEC project as a subcomponent of a communications monitoring toolset in an e-Commerce environment. The ENSURESEC project has received funding from the European Union’s Horizon 2020 research and innovation program. It aims to protect the whole range of modern e-Commerce by addressing a wide variety of threats. More specifically, it focuses on a wide variety of products ranging from virtual products and services purchased online to physical products bought online and delivered to the customers and aims at addressing numerous threats ranging from e-Commerce web applications attacks to frauds committed by customers or insiders, delivery issues, etc. [34 ].
The aforementioned toolset offers advanced monitoring capabilities aiming to ensure that the communication protocols in use, as well as the underlying communication infrastructure, function properly and safely. In this direction, certifiably correct verification methodologies (e.g., Decision Tree, Random Forest, KNN, Support Vector Machine, Voting Ensemble) are adopted, and various functionalities are included in the toolset. One of these functionalities is called Threat and Incident Detection and utilizes parts of the methodology described inSection 4 . Through this functionality, ENSURESEC users can identify malicious operations or threats at the network level by analyzing, filtering and matching semantically low-level events. Furthermore, they can gain insights into the structured relationships among the various types of items involved.
Advanced threats detection mechanisms are of vital importance for well-designed and secured e-Commerce systems in order to protect and ensure sensitive data that is being targeted by malicious users. Within this scope, the current study evaluates the performance of distinct ML algorithms inSection 5 and provides an additional countermeasure component (Threat and Incident Detection) by engaging two domains of high interest (i.e., ML and cybersecurity). This will help to enhance the analysis of threats patterns and learning from this process in order to detect, prevent and recognize similar types of attacks, enhancing the capabilities of cybersecurity teams to respond in real or near-real time to active types of cyber-attacks.
Five more functionalities were included in the so-called “Communication Monitor” toolset namely:
The Threat and Incident Detection functionality, together with the other functionalities of the Communication Monitor, are depicted inFigure 2 .
A part of the current research work has also been implemented in the context of the ENSURESEC project as a subcomponent of a communications monitoring toolset in an e-Commerce environment. The ENSURESEC project has received funding from the European Union’s Horizon 2020 research and innovation program. It aims to protect the whole range of modern e-Commerce by addressing a wide variety of threats. More specifically, it focuses on a wide variety of products ranging from virtual products and services purchased online to physical products bought online and delivered to the customers and aims at addressing numerous threats ranging from e-Commerce web applications attacks to frauds committed by customers or insiders, delivery issues, etc. [34 ].
The aforementioned toolset offers advanced monitoring capabilities aiming to ensure that the communication protocols in use, as well as the underlying communication infrastructure, function properly and safely. In this direction, certifiably correct verification methodologies (e.g., Decision Tree, Random Forest, KNN, Support Vector Machine, Voting Ensemble) are adopted, and various functionalities are included in the toolset. One of these functionalities is called Threat and Incident Detection and utilizes parts of the methodology described in
Advanced threats detection mechanisms are of vital importance for well-designed and secured e-Commerce systems in order to protect and ensure sensitive data that is being targeted by malicious users. Within this scope, the current study evaluates the performance of distinct ML algorithms in
Five more functionalities were included in the so-called “Communication Monitor” toolset namely:
A Threat Objects Fusion functionality through which users can fuse different objects into a unified object
A Similarity Degree Calculation feature that enables users to execute a character-by-character complex comparison algorithm among all the types of objects stored in or retrieved from a specific Knowledge Base, after the execution of certain processes
An Association Rule Engine that can be used for revealing hidden patterns and relations while exploring a populated database
A Visualization functionality that can be used for the interactive representation of populated ontologies in the form of graphs
An Advanced Reasoner through which users can apply rule-based logical reasoning into the existing Knowledge Base
The Threat and Incident Detection functionality, together with the other functionalities of the Communication Monitor, are depicted in
Free full text: Click here
