In the context of modern networks where cyber-attacks are increasingly complex and frequent, traditional Intrusion Detection Systems (IDS) often struggle to manage the vast volume of data and fail to detect novel attacks. Leveraging Artificial Intelligence, specifically Natural Language Processing with transformer architectures, offers a promising solution. This study applies the Bidirectional Encoder Representations from Transformers (BERT) model, enhanced by a Byte-level Byte-pair tokenizer (BBPE), to effectively identify network-based attacks within IoT systems. Experiments on three datasets-UNSW-NB15, TON-IoT, and Edge-IIoT-show that our approach substantially outperforms traditional methods in multi-class classification tasks. Notably, we achieved near-perfect classification accuracy on the Edge-IIoT dataset, with significant improvements in F1 scores and reduction in validation losses across all datasets, demonstrating the efficacy of pre-trained Large Language Models (LLMs) in network security.

This paper builds upon our previous work on Network Security Traffic Analysis Platforms (NSTAP) [1], presenting an advanced framework for the real-time monitoring of network traffic and endpoint security in large-scale enterprises. We employ a fully integrated technology stack that includes Elastic Stack, ZEEK, Osquery, Kafka, and GeoLocation data to create a comprehensive security analytics solution. A significant contribution of this research is the integration of supervised machine learning models into our platform, trained specifically on the UNSW-NB15 dataset. We explored three supervised machine learning algorithms - Random Forest (RF), Decision Trees (DT), and Support Vector Machines (SVM). For SVM, we also tested a dimensionality reduction algorithm to maximize model accuracy and optimized both computation time and performance. The evaluation is based on Accuracy, False Positive Rate (FPR) and revealed that the Random Forest Classifier, in conjunction with Pearson correlation-based feature selection methods, achieved the highest accuracy of 99.32% and an error rate of 0.67%.These findings not only substantiate the robustness of our unified platform but also set the stage for future research in developing scalable, efficient, and automated security solutions tailored for large enterprises.

Real-time traffic management and control have become necessary in today's networks due to their complexity and cybersecurity risks. With the increase in Internet use, threats are more prevalent and require real-time detection and analysis to prevent network intrusions. As the number of data flow increases, the number and the types of attacks increase, which makes detecting intrusions challenging. Therefore, over the last years, many researchers have focused on different ways to detect and more importantly prevent these intrusions. In this work, we describe the design and evaluation of a network security traffic analysis platform (NSTAP) that collects, searches, and analyzes traffic data in real time in order to filter out malicious flows. Through charts, tables, histograms, and other visualization methods, we demonstrate that the platform can produce powerful and useful insights with simple time-domain analytics of large data volumes. This work is intended to be the foundation for more automation tools based on machine learning.

By 2030, over a half trillion devices will be connected to the internet. With so many devices providing a wide range of features, there is a need for a framework for innovation and reuse of Internet of Things (IoT) and Cyber-Physical Systems (CPS) capabilities. Such framework should facilitate the composition of capabilities and provide stakeholders means to reliably model and verify compositions. An IoT and CPS Composition Framework (ICCF) is proposed to achieve this goal. ICCF is based on the NIST CPS framework composition guidelines, intuitive composition semantics inspired from the mPlane protocol, and strong formal verification capabilities of the Temporal Logic of Actions (TLA) formal descriptors and tools. This paper demonstrates why such framework, semantics, and formal specification and verification components form a powerful and intuitive composition framework that satisfies different stakeholders concerns. To achieve this purpose, semantics and formal specification of the composition algebra were provided, a well-being composite capability within a smart building was specified, its prototype model in a formal verification tool was run, an analysis of the results of symbolic execution quantitatively and qualitatively was performed, and assessment of the trustworthiness of the composition was done. Lastly, implementation details were provided and proposed extensions to other domains such as smart transportation and smart health were discussed.

Publicly available data is increasing rapidly, and will continue to grow with the advancement of technologies in sensors, smartphones and the Internet of Things. Data from multiple sources can improve coverage and provide more relevant knowledge about surrounding events and points of Interest. The strength of one source of data can compensate for the shortcomings of another source by providing supplementary information. Maps are also getting popular day-by-day and people are using it to achieve their daily task smoothly and efficiently. Starting from paper maps hundred years ago, multiple type of maps are available with point of interest, real-time traffic update or displaying micro-blogs from social media. In this paper, we introduce Hadath, a system that displays multi-resolution live events of interest from a variety of available data sources. The system has been designed to be able to handle multiple type of inputs by encapsulating incoming unstructured data into generic data packets. System extracts local events of interest from generic data packets and identify their spatio-temporal scope to display such events on a map, so that as a user changes the zoom level, only events of appropriate scope are displayed. This allows us to show live events in correspondence to the scale of view - when viewing at a city scale, we see events of higher significance, while zooming in to a neighbourhood, events of a more local interest are highlighted. The final output creates a unique and dynamic map browsing experience. Finally, to validate our proposed system, we conducted experiments on social media data.

Crowdsourcing offers great opportunities to recognise user context and prescribe relevant services for both offline and real-time activities. In this work, we present a zoning model that leverages spatio-temporal dimensions and then employs different contexts to recommend necessary customised services. The context model takes into consideration three context sets: fully restricted, fully unrestricted and semi-restricted with respect to both spatial and temporal dimensions. As a proof of concept, we apply this zoning model in a scenario where a very large crowd get together to perform spatio-temporal activities. The user context of the heterogeneous crowd is captured using the carried smartphones, i.e. via crowdsourcing. Depending on the context sets and zone, the system can recommend a set of services to each user. The system has been deployed since 2014 to support the spatio-temporal activities of a very large crowd. We present our implementation details and the user feedback, which is very encouraging.

International audience

Recommending an optimized path for a large crowd poses a unique challenge to existing routing algorithms due to the interactions between users and the dynamic changes over road networks. Industries, researchers and end users show an enormous interest in crowdsourced data comprising social networks and user-generated content to remain updated with their concerns. In this paper, we present a data collection framework that helps users to find optimized routes in a dynamic environment. We have developed a data collection framework to collect dynamic road conditions via a set of location-based services to support a very large Hajj crowd by capturing their locations using smartphones. We also collect geotagged social network data that provides more details about road conditions. The system leverages geotagged crowdsourced information to identify constraints such as accidents, congestions, and roadblocks. Moreover, by continuously collecting real-time geotagged data of moving users, the system can also find the flow of traffic and road conditions. We propose a spatial grid index to compute the optimized path, and to identify the affected users within impact zones. The plan is to test the whole application and back-end server during Hajj 2016, where over three million pilgrims from all over the world gather to perform their rituals.

Crowdsourced multimedia data poses several challenges when it is collected, stored, indexed, retrieved, and visualized. Examples of crowd source multimedia data are social sensors, vehicle sensors, physical sensors, human sensors, etc. Analyzing such multimodal and diversified crowdsourced data provides very rich understanding about the need of individuals within a crowd. Such understanding makes it possible to tailor services to individuals' needs, also called context-aware services. In this paper, we propose a spatial multimedia big data framework that can collect multimedia data from 1) a very large crowd equipped with multi-sensory smartphones, 2) vehicles, and 3) social networks. A set of multimedia services are offered to users to support their spatio-temporal activities. These include but not limited to 1) simple user interfaces to utilize multimedia services for instant guidance, 2) navigation to points of interests (POI), and 3) efficient and cost effective intra-city rides to users. The big data framework is designed to handle a very large number of multimedia spatio-temporal queries in real-time. The system is a pilot project and will be deployed during the event of Hajj 2015 when over three million pilgrims from all over the world will visit Makkah, Saudi Arabia to perform their Hajj rituals.

Social networks, such as Facebook, Instagram, and Twitter, provide intuitive ways to share a variety of information including geotagged multimedia data within users' communities of interest (COI) or publicly in real-time. Real-time geotagged multimedia data can provide semantics to conventional spatial queries in order to enrich the user navigation experience. In this paper, we introduce a mechanism to process spatio-temporal queries by leveraging geotagged multimedia data such as images, audio, video, and text, in order to add semantics to the conventional queries. Our framework collects, stores, and spatially tags multimedia data shared by users through social networks or through our developed mobile application. The system then uses such data in order to enhance the conventional routing services by resolving existing usability issues and by providing semantics to the routes in terms of enriched points of interest while taking dynamic road conditions into account. A proof of concept of the system will be demonstrated with the following spatio-temporal queries on road networks: 1) multimedia-enhanced shortest path queries; 2) multimedia-enhanced k-nearest neighbor queries; and 3) multimedia-enhanced range queries. Finally, a novel technique for finding lost individuals using geotagged multimedia data is also introduced. The results are tailor-made to the users' smartphone bandwidth and resolution requirements

Understanding crowd dynamics of large-scale events is crucial to deliver a pleasant experience for the participants. In this paper, we propose a novel hybrid participatory sensing approach to capture large group dynamics. Specifically, our approach is based on distributing a large number of tiny, wearable Bluetooth Low Energy (BLE) tags and few smartphones among the group members. We start by identifying the best configuration; in terms of transmit power and beaconing interval; for detecting BLE tags in indoor and outdoor environments. Then, as a case study, we deploy the system during the six main days of the Hajj pilgrimage in 2014, which is the world's largest annual religious gathering. We used the proposed hybrid participatory approach to collect the mobility data of pilgrim groups based on GPS location and co-occurring BLE tag detections. Our system provided up to 80% group-wise detectability in a single scan event. Moreover, 98% cumulative unique tags detectability is achieved in the whole event, leading to 0.74 million records. Analysis of the group dynamics revealed unexpected behavior and interesting findings. In particular, regional variations are found among different groups in the entry or exit times, duration of stay, and group cohesion; which are attributed to congestion, improper arrangements, or the nature of activities. Furthermore, we show sub-community clusters could be identified revealing clear biases based on the demographic characteristics. Based on this case study, we also give suggestions on using the proposed system for other large-scale events.

Traditional routing algorithms for calculating the fastest or shortest path become ineffective or difficult to use when both source and destination are dynamic or unknown. To solve the problem, we propose a novel semantic routing system that leverages geo-tagged rich crowdsourced multimedia information such as images, audio, video and text to add semantics to the conventional routing. Our proposed system includes a Semantic Multimedia Routing Algorithm (SMRA) that uses an indexed spatial big data environment to answer multimedia spatio-temporal queries in real-time. The results are customized to the users' smartphone bandwidth and resolution requirements. The system has been designed to be able to handle a very large number of multimedia spatio-temporal requests at any given moment. A proof of concept of the system will be demonstrated through two scenarios. These are 1) multimedia enhanced routing and 2) finding lost individuals in a large crowd using multimedia. We plan to test the system's performance and usability during Hajj 2015, where over four million pilgrims from all over the world gather to perform their rituals.

This paper presents a system that enables efficient and scalable real-time user and vehicle discovery using textual, audio and video mechanisms. The system allows users to group together for shared intra-city transportation with the aid of multimedia that helps individuals to 1) find community of common interest (CoCI), 2) locate individual users in a large crowd and 3) locate vehicles for mobility in an efficient and cost effective manner. The system is a pilot project and will be deployed during Hajj 2015 when over three million pilgrims from all over the world visit Makkah, Saudi Arabia.

We propose a context aware framework that offers a set of cloud-based services to support a very large Hajj and Umrah crowd by capturing their contexts using smartphones. The proposed framework captures the individual's context, provides a set of adapted services, and allows being in touch with a subset of one's community of interest. We leverage the spatiotemporal sensory data captured by our framework to define users' contexts for optimized services. Our proposed framework is also envisioned to assist the Hajj and Umrah authorities to (1) improve Hajj & Umrah documentation, (2) improve Hajj organization through better understanding of pilgrims' (individual and crowd) spatial and temporal behavior and needs, and (3) protect pilgrims' environment through environmental monitoring. In particular, the developed methods, techniques, and algorithms will support the pilgrimage quality of experience. We have tested our system through end-user subjects and due to apply for the upcoming Hajj events. We present our implementation details and the general impression of end users about our system.

In this paper, we present a novel idea of a map-based therapy environment for people with Hemiplegia. The therapy environment is designed according to the suggestions of therapists, which consists of a spatial map browsing serious game augmented with our novel multi-sensory natural user interface (NUI). The NUI is based on 3D motion sensors that can recognize different hand and body gestures used for browsing a 3D or 2D map. The 3D motion sensors work in a non-invasive way; hence, they do not require any wearable body attachments and can be used at home without assistance from the therapists. The map-browsing environment provides an immersive experience to the disabled users, which helps in performing therapy in an interesting and entertaining manner. We have developed analytics for measuring certain quality of health improvement metrics from each type of spatial map browsing movements. The 3D motion sensors have been tested with Nokia, Google, ESRI, and a number of other maps that allow a subject to visualize and browse the 3D and 2D maps of the world. The map browsing session data shows the nature of big data; hence, the session data is stored in a cloud environment. Our developed serious game environment is web-based; thus anyone having the appropriate low cost sensor hardware can plug it in and start experiencing a natural way of hands free map browsing. We have deployed our framework in a hospital that treats Hemiplegic patients. Based on the feedback obtained, the developed platform shows a huge potential for use in hospitals that provide physiotherapy services as well as at patients' home as an assistive therapeutic service.