This paper describes the development and implementation of a natural language processing model based on machine learning which performs cognitive analysis for cybersecurity-related documents. A domain ontology was developed using a two-step approach: (1) the symmetry stage and (2) the machine adjustment. The first stage is based on the symmetry between the way humans represent a domain and the way machine learning solutions do. Therefore, the cybersecurity field was initially modeled based on the expertise of cybersecurity professionals. A dictionary of relevant entities was created; the entities were classified into 29 categories and later implemented as classes in a natural language processing model based on machine learning. After running successive performance tests, the ontology was remodeled from 29 to 18 classes. Using the ontology, a natural language processing model based on a supervised learning model was defined. We trained the model using sets of approximately 300,000 words. Remarkably, our model obtained an F1 score of 0.81 for named entity recognition and 0.58 for relation extraction, showing superior results compared to other similar models identified in the literature. Furthermore, in order to be easily used and tested, a web application that integrates our model as the core component was developed.

The aim of this paper was to enhance the process of diagnosing and detecting possible vulnerabilities within an Internet of Things (IoT) system by using a named entity recognition (NER)-based solution. In both research and practice, security system management experts rely on a large variety of heterogeneous security data sources, which are usually available in the form of natural language. This is challenging as the process is very time consuming and it is difficult to stay up to date with the constant findings in the areas of security threats, vulnerabilities, attacks, countermeasures, and risks. The proposed system is conceived as a semantic indexing solution of existing vulnerabilities and serves as an information tool for security management experts. By integrating the proposed system, the users can easily discover the potential vulnerabilities of their IoT devices. The proposed solution integrates ontologies and NER techniques in order to obtain a high rate of automation with the scope of reaching a self-maintained and up-to-date system in terms of vulnerabilities and common exposures knowledge. To achieve this, a total of 312 CVEs (common vulnerabilities and exposures) specific to the IoT field were identified. CVEs are arguably one of the most important cybersecurity resources nowadays, containing information about the latest discovered vulnerabilities. This set is further used as data corpus for an NER model designed to identify the main entities and relations that are relevant to IoT security. The goal is to automatically monitor cybersecurity information relevant to IoT, and filter and present it in an organized and structured framework based on users’ needs. The taxonomies specific to IoT security are implemented via a domain ontology, which is later used to process natural language. Relevant tokens are marked as entities and the relations between them identified. The text analysis solution is connected to a gateway which scans the environment and identifies the main IoT devices and communication technologies. The strength of the approach proposed within this research is that the designed semantic gateway is using context-aware searches in the modeled IoT security database and can identify possible vulnerabilities before they can be exploited.

This article describes a process for introducing modern technological subjects into the academic curricula of non-technical universities. The process described may increase and contribute to social sustainability by enabling non-technical students’ access to the field of the Internet of Things and the broader Industry 4.0. The process has been defined and tested during a curricular reform project that took place in two large universities in Eastern Europe. In this article, the authors describe the results and impact, over multiple years, of a project financed by the European Union that aimed to introduce the following subjects into the academic curricula of business students: cloud computing, big data, mobile programming, and social networks and cybersecurity (CAMSS). The results are useful for those trying to implement similar curricular reforms, or to companies that need to manage talent pipelines.

Research and development activities are one of the main drivers for progress, economic growth and wellbeing in many societies. This article proposes a text mining approach applied to a large amount of data extracted from job vacancies advertisements, aiming to shed light on the main skills and demands that characterize first stage research positions in Europe. Results show that data handling and processing skills are essential for early career researchers, irrespective of their research field. Also, as many analyzed first stage research positions are connected to universities, they include teaching activities to a great extent. Management of time, risks, projects, and resources plays an important part in the job requirements included in the analyzed advertisements. Such information is relevant not only for early career researchers who perform job selection taking into account the match of possessed skills with the required ones, but also for educational institutions that are responsible for skills development of the future R&D professionals.

The article studies the current text processing tools based on Artificial Intelligence. A literature review is done emphasizing the dynamic evolution of AI-powered text analytics, having as its central tool ChatGPT and its capabilities. The focus is centered on the techniques and methods that are using embeddings in order to improve large language models (LLMs). In this paper is analyzed the current situation of the literature in terms of text processing using Retrieval-Augmented Generation and is highlighted the potential of this technology to enhance the interpretability and trust in applications critical, such as those related to education or business. AI has revolutionized natural language processing (NLP), which facilitated the machines to interpret and generate text efficiently and accurately. In addition, large language models with external knowledge bases have been developed. These are used to produce more accurate and contextually relevant text responses. This approach is called Retrieval-Augmented Generation (RAG is one of the most significant recent advancements in this field. Based on our study, two use cases are implemented to show the applicability of our study: one related to education and one related to business IT-related documents. The methodology describes the techniques used. This includes retrieval-augmented generation and embedding stored using vector databases. Our custom models are evaluated by comparing them to the general ones, without embeddings, showing superior performance. The article highlights remarkable progress in Retrieval-Augmented Generation (RAG), which is used for AI text processing with a focus on business and education fields. Further in this paper, many of the most significant highlights are presented, which include a scalable framework for AI applications, a new integration of Retrieval-Augmented Generation and embeddings, practical application demonstrations, bridging gaps in the analysis op AI text, significant development in AI performance and optimizing educational and business processes.

This article analyzes and highlights the security perspective of Internet of Things (IoT) connected devices and their communication challenges, as IoT is considered one of the key emerging fields in Industry 4.0. The IoT architectures can consist of physical systems, virtual ones or even hybrids, combining a collection of different physically active things, sensors, cloud services, specific IoT protocols, communication layers, users and developers. On top of all, it is the business layer, because the scope of the entire IoT environment is to deliver data, to monitor and to facilitate the management of complex processes. In order to facilitate the data exchange between the IoT layers, there have been developed a series of protocols particular to the IoT domain. As in many IT related fields, the solutions are not perfect from the data security and privacy perspectives, many challenges being still open research issues. As the two concepts of IoT and Cloud of Things are connected, bringing real world data into the Cloud to process it, raises Cloud Computing security concerns regarding the privacy and security of data. Although in recent years, many efforts have been made to improve Cloud Computing security, there are risks that need to be taken into consideration. From the Web of Data’s point of view, things are even more prone to security risks. Because privacy is one of the fundamental right of digital users, it is extremely important for new technologies to comply with privacy regulations and policies, such as the new European data protection and privacy frameworks. In this context, companies must take into account standards, challenges and new trends in IoT. In the absence of specific measures, raw or processed data can be easily stolen from the Web of Data. In this paper we analyze and present the main protocols of communication in the IoT field from a data security perspective. Also, we do a review of the main architectures that can improve the security of the communication between IoT devices and the Cloud data storage.