A careful analysis of the Large Language Model (LLM) results, generated through anonymized representations of the original dataset, is crucial to precisely evaluate the data-sharing procedure's limitations and facilitate valuable collaborations among Internet-based cognitive behavioral therapy (ICBT) companies and third parties. This paper presents an experimental study of fine-tuning 27 LMs for a multiclass classification task to identify depression severity using 40,191 tweets labeled by human annotators. We fine-tune 14 Bidirectional Encoder Representations from Transformers (BERT), 6 Robustly Optimized BERT Pretraining Approaches (RoBerta), 3 Generative Pretraining (GPT), and 4 Text-to-Text Transfer Transformer (T5) based LMs to classify confidential and anonymized tweets. We report that T5, through conditional generation, outperforms widely adopted BERT, RoBerta, and GPT types for classifying confidential and anonymized tweets. Anonymizing personal information safeguards user privacy and often increases LM performance. Case sensitivity can potentially improve or harm the performance of domain-specific LMs for original and anonymized text.

Conversational agents, powered by Large Language Models (LLMS) have seen significant advancements, enabling them to produce coherent and contextually relevant text. However, a major challenge remains in the form of Hallucination, where the LLMs generate content that is fabricated and factually incorrect. This issue is especially critical in domains that require high factual accuracy, where it is essential to ensure the reliability and trustworthiness of LLM-generated content. This paper provides a novel metric based method towards detection of hallucinations in natural language generation. The method combines three complementary components: Jaccard similarity for entity-level alignment, natural language inference for semantic consistency, and a normalized perplexity measure for fluency. The integration of these components through a weighted average produces a hallucination metric where a high value indicates a potential occurrence of hallucination. This metric increases the reliability of LLM-generated content. The method’s effectiveness is evaluated on benchmark datasets (BEGIN and FEVER), which demonstrates improved performance in hallucination detection. The evaluation results indicate that the proposed method provides a balanced approach to identifying hallucinations, offering a promising solution for more trustworthy conversational agents.

Generative AI and large language models (LLMs) offer a new type of decision support that permits the user to interact with the system using natural language. LLMs are particularly useful for pragmatic reasons. While syntax and semantics have been easier to handle over the years, pragmatics has been difficult to manage, at least before the invention of LLMs. Users interact with the system through one or more prompts to obtain a response. ‘Prompt engineering’ is the art and science of designing and refining effective prompts to elicit the desired output from the LLM. Prompt engineering suggests techniques to design inputs, acquire data and information, and grasp the capabilities and limitations of the models. Prompts contain questions or instructions which are trimmed to optimize the prompts and obtain the desired answers, i.e., changing different parameters to extract the required data. These techniques range from trivial prompts to knowledge-intensive prompts. This paper explores prompt engineering directed towards supporting human decision making and suggests a framework for using LLMs towards decision goals. An overview of prompting techniques is provided with benefits and drawbacks along with examples. In addition, an overview of decision support is presented with a focus on the intelligence, design, choice and implementation paradigm. The paper proposes an LLM/DM framework for prompting for decision support with current LLMs and presents an empirical example.

Natural language processing, with parsing and generation, has a long tradition. Parsing has been easier to perform than a generation but with generative artificial intelligence (a.k.a Gen AI) and large language models (abbr. LLMs), this has changed. Generative artificial intelligence is a type of artificial intelligence that uses a large data set to create something in the genre of that data set. It can generate different outputs ranging from texts, audio, objects, pictures, and paintings to videos, but also synthetic data. LLMs use deep learning and deep neural networks to train on large text corpora for recognizing and generating texts. These models are based on massive data sets, collected from databases and the web. They use transformer models to detect how elements in sequences relate to each other. This provides context support. Two well-known large language models are the Generative Pre-trained Transformer, GPT, used in ChatGPT and Bidirectional Encoder Representations from Transformers, BERT. Although LLMs have advantages, they have problems. This paper presents generative artificial intelligence and LLMs with benefits and drawbacks. Results from applying these models have shown that they can work well for accuracy in specificity, user personalization and human-computer communication but they may not provide acceptable, reliable and truthful results. For example, ethics, hallucinations and incorrect information, or misjudgments, are some major problems. The paper ends with future directions, research questions on LLMs, and recommendations.

Deep Learning has the ability to train on datasets created from videos. This facility makes deep learning algorithms suitable for detecting distinct objects in large sets of frames, particularly for delineating anomalies like precancerous lesions during surveillance colonoscopies of the large bowel. However, capturing subtle, diffuse characteristics in these videos’ frames can be a challenge. This paper presents a deep learning system that uses colour channel separation and recombination of images to improve the performance of an object detection model, to tackle this challenge. Using a dataset from surveillance colonoscopy videos to find precancerous and cancerous lesions in IBD patients, individual colour channels of RGB images in the dataset are separated and recombined to form different datasets which are later used to train YOLOv8x models. The object detection model that is trained and tested with only the blue channel component of the images in the dataset performs better and gives more accurate predictions than the object detection model that is trained and tested with the datasets containing the green and/or the red channel images, as well as the dataset containing the original RGB images.

Individual autonomous agents can employ adaptive Artificial intelligence. The individual agent can respond fast, near real-time, to the environment and avoid hurdles and collisions while keeping a certain distance. Although the systems learn to handle the environment with new conditions and situations, it does not, per see, guarantee that the system is robust meaning the system performs predictably while its variables and assumptions are altered. Nor does it mean that adaptive individual autonomous agents in multi-agent systems perform well in dynamic, distributed and partially observable environments where unexpected things happen. This paper presents an approach to applying robustness analysis on adaptive Artificial intelligence in multi-autonomous agent systems. Adaptiveness modifies the agents’ learning by generating appropriate responses to new situations to achieve resilience to perturbations. The robustness analysis explores the quality of these responses so the agent can continuously operate despite abnormalities in input and consequently safely tolerate perturbations. A multi-agent prototype simulating the integration of adaptive agents with robustness analysis shows that it is possible to apply robustness analysis to the responses of adaptive Artificial intelligence. In the system, each autonomous agent collects data about the surrounding environment and applies adaptation to provide options, i.e., possible actions and decisions. The robustness analysis examines the options in order to validate and adjust them as needed. With these updated options, the agents retrain to resist malfunctioning and achieve resilience and robustness for given situations.

Autonomous, mobile cyber-physical systems are becoming popular in the transportation, manufacturing and industry sectors, thanks to their ability to provide smart and twenty-four-seven services to society, production industries and companies. A smart service is, for example, autonomous transportation of people in traffic or goods in manufacturing industries without human control. The use of autonomous, mobile cyber-physical systems has led to changes and challenges. Especially, dynamic obstacles are affecting the cyber-physical systems and these must be handled to provide robust environments. Although the mobile cyber-physical systems are interacting and acting with the other moveable cyber-physical systems, they also need to handle interactions with human beings. These human beings' interactions can be verbal conversations with the system. However, this communication may struggle with language barriers where a cyber-physical system, for example, only works with one or two natural languages. These interaction limitations may not be obvious nor even known to human beings. In addition, the human being may not be able to handle verbal communication with the cyber-physical systems, therefore hand signs are required to communicate with the systems to provide a trustworthy and secure environment together with several cyber-physical systems. This paper presents Handie, a system handling hand sign interactions with autonomous mobile cyber-physical systems, aka mobile robots. The hand signs are simple, i.e., signs for "ok", "thumbs-up" and "stop" that constitute commands to the robot. Moreover, this non-verbal interaction includes human moods, with facial expressions like happiness, surprise, and fear. Both hand signs and facial expressions are handled by a deep learning recognition component in a humanoid, programmable robot Softbank Robotics' NAO robot v6. Results from testing the system show that the cyber-physical system can recognize hand signs and facial expressions to an acceptable extent to be useful, although the interaction is time-consuming and slow. Also, currently, the communication back to the end users is made verbally and not visually. The next step of Handle's will be to expand the number of hand signs and improve the interaction with the NAO robot.

A key driver to ofering smart services is an infrastructure of Cyber-Physical systems (CPS)s. By definition, CPSs are intertwined physical and computational components that integrate physical behaviour with computation. The reason is to autonomously execute a task or a set of tasks providing a service or a list of end-users services. In real-life applications, CPSs operate in dynamically changing surroundings characterized by unexpected or unpredictable situations. Such operations involve complex interactions between multiple intelligent agents in a highly non-stationary environment. For safety reasons, a CPS should withstand a certain amount of disruption and exert the operations in a stable and robust manner when performing complex tasks. Recent advances in reinforcement learning have proven suitable for enabling multi-agents to robustly adapt to their environment, yet they often depend on a massive amount of training data and experiences. In these cases, robustness analysis outlines necessary components and specifications in a framework, ensuring reliable and stable behaviour while considering the dynamicity of the environment. This paper presents a combination of multi-agent reinforcement learning with robustness analysis shaping a cyber-physical system infrastructure that reasons robustly in a dynamically changing environment. The combination strengthens the reinforcement learning, increasing the reliability and flexibility of the system by applying robustness analysis. Robustness analysis identifies vulnerability issues when the system interacts within a dynamically changing environment. Based on this identification, when incorporated into the system, robustness analysis suggests robust solutions and actions rather than optimal ones provided by reinforcement learning alone. Results from the combination show that this infrastructure can enable reliable operations with the flexibility to adapt to the changing environment dynamics.

Real-world autonomous systems are often controlled using conventional model-based control methods. But if accurate models of a system are not available, these methods may be unsuitable. For many safety-critical systems, such as robotic systems, a model of the system and a control strategy may be learned using data. When applying learning to safety-critical systems, guaranteeing safety during learning as well as testing/deployment is paramount. A variety of different approaches for ensuring safety exists, but the published works are cluttered and there are few reviews that compare the latest approaches. This paper reviews two promising approaches on guaranteeing safety for learning-based robust control of uncertain dynamical systems, which are based on control barrier functions and control Lyapunov functions. While control barrier functions provide an option to incorporate safety in terms of constraint satisfaction, control Lyapunov functions are used to define safety in terms of stability. This review categorises learning-based methods that use control barrier functions and control Lyapunov functions into three groups, namely reinforcement learning, online and offline supervised learning. Finally, the paper presents a discussion of the suitability of the different methods for different applications.

Autonomous cyber-physical systems, CPS, in dynamic environments must work impeccably. The cyber-physical systems must handle tasks consistently and trustworthily, i.e., with a robust behavior. Robust systems, in general, require making valid and solid decisions using one or a combination of robust reasoning strategies, algorithms, and robustness analysis. However, in dynamic environments, data can be incomplete, skewed, contradictory, and redundant impacting the reasoning. Basing decisions on these data can lead to inconsistent, irrational, and unreasonable cyber-physical systems’ movements, adversely impacting the system’s reliability and integrity. This paper presents the assessment of robust reasoning for autonomous cyber-physical systems in dynamic environments. In this work, robust reasoning is considered as 1) the capability of drawing conclusions with available data by applying classical and non-classical reasoning strategies and algorithms and 2) act and react robustly and safely in dynamic environments by employing robustness analysis to provide options on possible actions and evaluate alternative decisions. The result of the research shows that different common existing strategies, algorithms and analyses can be provided together with a comparison of their applicabilities, benefits, and drawbacks in the context of cyber-physical systems operating in dynamically changing environments. The conclusion is that robust reasoning in cyber-physical systems can handle dynamic environments. Moreover, combining these strategies and algorithms with robustness analysis can support achieving robust behavior in autonomous cyber-physical systems while operating in dynamically changing environments.

This paper aims at identifying the key role of ”Smart Energy and Power Systems Modelling”, within the context of Energy Informatics. The main objective is to describe how the specific subject of ”Smart Energy and Power Systems Modelling” can give a key contribution within the novel domain of Energy Informatics, by successfully linking and integrating the different disciplines involved. First the paper will present how and where the specific subject of ”Smart Energy and Power Systems Modelling” can position itself within the broad Energy Informatics domain. Afterwards the paper will explain how Cyber-Physical Systems (CPS) and Information and Communication Technologies (IoT), coupled with ”Smart Energy and Power Systems Modelling”, can enhance the Energy Informatics domain. In addition, the main challenges and opportunities for Energy Informatics specialists will be outlined, with regard to the interdisciplinary approach that characterises this field.

Information and Communication Technology (ICT)-infrastructures are increasingly important for enabling technology within smart society with smart cities and communities. An ICT-infrastructure handles data and information and encompasses devices and networks, protocols and procedures including Internet, Internet of Things and Cyber-Physical Systems. The current challenges of ICT-infrastructures are delivering services and applications that are requested by users, such as residents, public organisations and institutions. These services and applications must be combined to enhance and enrich the environment and provide personalised services. This requires radical changes in technology, such as dynamic ICT-infrastructures, which should dynamically provide requested services to be able to build smart societies. This paper is about pursuing smart and connected cities and communities by creating smart volatile ICT-infrastructures for smart cities and communities, called Ipsum. The infrastructure is of multidisciplinary art and includes different kinds of hardware, software, artificial intelligence techniques depending on the available parts and the services to be delivered. The goal is to provide a powerful and smart, and cost-saving volatile ICT-infrastructure with person-centred, ubiquitous and malleable parts, i.e., devices, sensors and services. Volatile means in real-time constitute a volatile network of devices and deploying it into cities and communities. Ipsum will be smart everywhere by collaborating with several different hardware and software systems and cooperating to perform complex tasks. By including ubiquitous and malleable parts in the infrastructure, Ipsum can facilitate an informed and engaged populace.

High performance sensor-based systems require distributed methods that quickly release threads and allow hardware to scale as the amount of data increases. Each sensor, component, has to handle a massive amount of data, independently, at the same time it must cooperate with other sensors, components. To facilitate rapid data processing between different components and handle the dataflow simultaneously, knowledge about the components, i.e., metadata, can be incorporated. This paper presents a sensor-based system that applies metadata on internal components and utilizes disaggregation and processing as core services to handle large amount of data from various distributed sensors. By combining components with individual advantages, the system can be designed to allow a high amount of simultaneous data to be disaggregated into the respective categories, which are processed to make information accessible and stored in a database that can easily be accessed by interested parties. To evaluate the system, performance tests have been carried out using metadata on combined components, disaggregation and processing. The tests show that it is possible to build faster and more reliable sensor-based systems that are scalable, fault tolerant and high performing.

Cyber-physical systems are integrations of computerized physical things in the environment that are merged by communications, and computations using embedded systems and networks. To make the components of the systems act intelligently, according to users’ needs, and understand and predict behavior of the cyber-physical systems, the cyber-physical systems need reasoning to link the outcome of the different components to be able to make use of the devices in the surrounding environment. The reasoning includes collecting sensor data, finding key concepts in the data and drawing conclusions that cyber-physical systems can use to control the components surrounding the users. However, there is a range of different components in a system where each has particular communication input-output and its own set tasks, which provides particular challenges associated with controlling or predicting the behavior of such systems, which require a kind of analytic tools. This paper presents reasoning strategies for smart cyber-physical systems that can extract and combine data, information, and knowledge to provide an intelligent behavior from users point of view. The reasoning strategies use users’ needs as a starting point and provide an environment that gives personalized support.

This paper is a philosophic look at(view of) the promise and perils of cyber physical systems. It is not a doom and gloom view of the systems (It is overview of the current systems and their technologies), we believe in the bright and useful future of the cyber physical environments (systems?). However, looking at the history of technology in the recent past, the need of careful prescription to design systems that prevent the unforeseen less than positive side of technology. No matter how useful a technology is, the providers soon push burdensome “features” on to the unsuspecting user.

The world is moving towards a time where more and more objects, like Internet of Things, will be connected in cyber-physical systems. These objects need excellent tools to collect information from sources in the surrounding environment. By using the wireless communication technologies of modern smartphones, such as Bluetooth, Near Field Communication, and Wi-Fi, a solid ground to transmit and receive information to and from various sources can be established. However, an obstacle communication protocols are needed to make the different devices transmit information. This paper presents a communication protocol for cyber-physical systems using wireless technologies and cloud computing to facilitate information exchange between objects. By constructing a communication protocol and implementing a system on top of the protocol, users can exchange useful information with the help of a smartphone or other devices with similar functionalities in cyber-physical systems. The protocol should enable using services with ease, hiding the complex underlying structure and make interactions as natural as possible.

Cyber-physical systems are often developed with an emphasis on the network of computational elements and the linkage between the computational and physical elements. The physical elements are different kinds of Internet of Things devices that carry out desirable and valid tasks from instructions. However, due to the limitations of current individual-based secure products and delivery of services, the requisites of these products and services have started to increase and, hence, the requirements for intelligent automated, networked and mobile devices arise. The current state of communication between the elements in Internet of Things is data exchange and needs step up to next level to improve the interaction with the surrounding devices to augmenting human capabilities. This paper presents an infrastructure for individualised intelligent decision-making and negotiation in cyber-physical systems with smart Internet of Things devices. The decision-making and negotiation is based on individual preferences to provide the best individual-based solutions. The solution is applied to health care, which will permeate throughout the paper.

Human intelligence draws its conclusions from a base of experience-generated knowledge. Beside being able to use this knowledge, it is limited to how much can be accessed at a time and this reasoning is often shown to be illogical, with respect to mathematical logics. The human's knowledge is always growing and being modified by current experience. In addition, the humans’ processing capability appears to be severally limited. This limitation is far from being a burden; it is part of the brilliance of the solution. The human mind does a every effective job of dealing with the world. For proof, we invoke the fact that human kind as survived and thrived. The current work is a first step in exploring a reasoner than can act in a human inspired performance, which is in the direction of general artificial intelligence.

Metadata, such as the domain description and the purpose of an ontology, can be used to describe the context of ontologies for ontology integration. However, these metadata are not always available in ontologies. To solve the problem, a method is developed to automatically discern the domain of an ontology. This method uses a so-called core domain ontology, rules and an ontology reasoner to identify the domain. The core domain ontology is a light weight ontology that consists of the essential concepts of a domain. Rules and the ontology reasoner are used to test if the core domain ontology is consistent with an ontology for which the domain needs to be identified. If the two ontologies are not in violation, then the method confirms the consistency between them, that is, the test ontology shares the same domain as the core domain ontology. If the core domain ontology shows inconsistency with the test ontology, they do not share the same domain and then the ontology can be used to compare with another core domain ontology. Experiments on the core domain ontology for the conference domain show good results. Ten ontologies of mixed domains are compared with the core conference ontology. Eight ontologies’ domain are correctly identified, out of which, four ontologies are identified as sharing the same ontology domain.

AI-complete systems developed today, are commonly used for solving different artificial intelligence problems. A problem is a typical image recognition or speech recognition, but it can also be language processing, as well as, other complex systems dealing with general problem solving. However, no AI-complete system, which models the human brain or behavior, can exist without looking at the totality of the whole situation and, and hence, incorporating an AI-computerized sensory systems into a totality that constitute a combination of senses. This paper proposes a combination of sensory systems to form a comprehensive AI-system by combining the different senses, called AIC –AI-system for a combination of senses. The AIC-system is not a complete system in the sense that it contains a total set of information or uses all kinds of digital sensory systems. Nonetheless, it is a system under self-development. It develops its own knowledge base, as experiences, which will be based on the different characteristics: images, sounds, smells, tastes, touches with emotions/feelings and expressions. The result is a kind of perception of the surrounding environment.

Software security engineering requires an understanding of the security issues and knowledge about how to solve these issues. Unfortunately, the engineers often lack knowledge in security field, which induces security risks in software systems. To minimize the risks and support engineers during system development, structured and reusable information in security area is required. To this objective, security process and security patterns for software development are proposed. Moreover, the design of the security patterns is based on ontology techniques, which can provide structured information that can be reused and combined. For searching and mapping of patterns, we use agents in multi-agent system. The presented approach can enhance understanding of security issues and support implementation of security in software engineering process.

By integrating ontologies, knowledge represented in these ontologies can be extended and reused for purposes like building new ontologies, composing advanced services on the semantic web and sharing knowledge. The ontology integration is a difficult task because of semantic barriers. In this paper, an approach of building context for ontology integration is presented. The integration of the ontologies is carried out by using context and ontology violation check. To provide context for the integration, context rules are utilized for interpretation and reasoning across ontologies. The ontology violation check is applied in the process of building ontology intersection under the open world assumption. As a result of the integration, an ontology intersection is produced, which is larger than the intersection of entities, but not larger than the union of the original entities.

Reverse engineering, also called reengineering, is used to modify systems that have functioned for many years, but which can no longer accomplish their intended tasks and, therefore, need to be updated. Reverse engineering can support the modification and extension of the knowledge in an already existing system. However, this can be an intricate task for a large, complex and poorly documented knowledge-based system. The rules in the knowledge base must be gathered, analyzed and understood, but also checked for verification and validation. We introduce an approach that uses reverse engineering for the knowledge in knowledge-based systems. The knowledge is encapsulated in rules, facts and conclusions, and in the relationships between them. Reverse engineering also collects functionality and source code. The outcome of reverse engineering is a model of the knowledge base, the functionality and the source code connected to the rules. These models are presented in diagrams using a graphic representation similar to Unified Modeling Language and employing ontology. Ontology is applied on top of rules, facts and relationships. From the diagrams, test cases are generated during the reverse engineering process and adopted to verify and validate the system.

Leadership requires making decisions and implementing the results by influencing those being lead. The Personal Access to Leadership project, PAL, is constructing tools to assist leaders in creative ways and assisting the development of leaders. The tools are knowledge-based systems employing shallow understanding of the domain. The approaches used provide guidance, but do not generate solutions. One aspect that continues under exploration in PAL is the use of stories for training and guiding leaders. In order to make such support systems work, a representation is needed to enable locating useful stories related to the task of a leader. This article defines a model for story representation for the PAL tools, called the PAL tool IdeaLab. The IdeaLab is the tool to which stories are being added, as a help system to support and extend the user's thinking.