Abstract Following the period marked by the spread of the COVID-19 virus, tourists and enthusiasts flocked to museums and archaeological parks, highlighting persistent and emerging challenges concerning the enhancement of cultural heritage. This study proposes a novel framework that leverages context-aware recommender systems (CARS) to personalize the cultural experience based on the contextual conditions in which it occurs. The framework has been specifically developed and tested within the archaeological sites of Paestum and Pompeii. The evaluation process was conducted in two phases. The first phase focused on assessing the accuracy and reliability of the proposed context-aware approach, using a dataset comprising 972 user ratings collected from visitors to the Pompeii archaeological park. The second phase measured user satisfaction through a large-scale, in situ evaluation conducted directly within the archaeological parks. This phase involved nearly 2000 participants and aimed to assess the real-world usability and perceived satisfaction of the platform through direct interaction in cultural settings. The results confirm the effectiveness of the approach, with over 90% of respondents expressing a high level of satisfaction.

The growing intricacy in the administration of solar systems necessitates novel strategies to guarantee efficiency and reliability. This research presents a sophisticated methodology for developing a Digital Twin for the solar system at the University of Salerno campus. The primary aim is to create a dynamic digital model that facilitates real-time monitoring of the system's operational conditions and enhances predictive maintenance techniques. The integration of sophisticated technologies, including the Internet of Things (IoT), Building Information Modeling (BIM), and Geographic Information Systems (GIS), enabled the creation of a virtual replica of the physical system. The ThingsBoard platform centralizes and analyzes the real-time data collected by IoT sensors, while Autodesk Revit and Dynamo create a parametric BIM model that enables a comprehensive and interactive depiction of the system. Georeferencing and spatial analysis, executed using ArcGIS, enhance the Digital Twin, facilitating a thorough comprehension of the interactions between the system and its environmental surroundings. The findings indicate that the Digital Twin facilitates prompt anomaly identification, failure prevention, and the simulation of predictive maintenance scenarios. This approach enabled the optimization of operational management and a substantial reduction in maintenance expenses. The method shows how the Digital Twin can be used to effectively manage solar power plants and can serve as a model for managing other important infrastructures. This helps promote sustainable and resilient management.

Abstract Protecting cultural heritage (CH) is a strategic activity for all countries, like Italy, which has many ancient properties that can be degraded and damaged over time. Archaeological sites are critical CH assets for Italy, and their management and protection are crucial. Visitors can intentionally or unintentionally damage archaeological sites, while natural events like rain, wind, sun, and weeds can degrade or damage such CH assets. In such a context, modern technologies can effectively support monitoring activities. This paper presents the design of a framework for acquiring aerial images and their analysis, supporting operators on a site with detailed maintenance suggestions and information, and allowing the launch of new precision surveys to investigate identified issues better. We also propose a prototype tool for automatically detecting maintenance issues in an archaeological site based on AI models applied to aerial orthophotos of the site. The case study taken in the exam is related to the archaeological site of Pompeii, which provides high-definition orthophotos of its artistic resources using aerial drones. A prototype tool is proposed to discover such maintenance issues in such images rapidly, present them straightforwardly to support human operators’ decision-making and understand which site zone needs more attention. The maintenance issues to identify fall into four classes: weedy vegetation, damaged conduits, damaged structures, and broken tiles. In the experimental phase, a custom dataset was used to train and evaluate various versions of Yolo model. The best performance has been obtained through the YoloV5l detector, with a F1 score of 0.482, 0.427 for mAP0.5 and 0.264 for mAP0.5–0.95 on cross-validation, a 0.502 F1-score, 0.482 mAP50 and 0.279 mAP50-95 on the test set. The model has a FPS capacity of 54.945 frame/sec. The tool has proven good efficiency being capable of scanning and analyse an entire orthophoto of about 10 GBs in a few minutes.

The employment of Information and Communication Technologies helps Cultural Heritage enhancement and allows users’ cultural experience improvement. In particular, multimedia content employment enables better involvement through Storytelling techniques aimed at capturing the users’ attention. Moreover, to guarantee the personalization of the cultural experience, Digital Storytelling requires integrating tools for identifying the users’ preferences in the specific environment in which the cultural experience evolves. Therefore, this paper aims to introduce an architecture in which a Context-Aware Recommender System is able to provide personalized paths in which each Point of Interest is presented according to Digital Storytelling techniques. The proposed architecture is validated via two experimental phases: the first aimed to evaluate the accuracy of the recommender system, and the second based on user experience. Obtained results are promising.

Spa treatments may mistakenly be considered palliative compared to traditional medicines; however, this is not the case. Mineral/thermal waters are medicines for all intents and purposes and should be analyzed and used as such. The difference in spa treatments compared to other medicines is the greater complexity with which they are delivered. Patients must follow a course of treatment that can last up to a couple of weeks, during which the effects of the therapy gradually go into evidence. Both inside and outside the spa facility, having patient monitoring could be a valuable tool to measure the effectiveness of treatment and possibly even intervene with personalized care based on the parameters detected. New technologies and paradigms such as the Internet of Things can offer a valuable tool to improve spa care through active monitoring of patients, both inside and outside the facilities, by going to measure what are the key parameters (i.e., heart rate, blood oxygenation, etc.) to track the progress of the therapy accurately and precisely during treatment. In particular, wearable devices (smartwatches or smart bands) can perform constant and non-invasive monitoring of the patient's status and the therapy itself. Therefore, the work aims to define a framework based on the Internet of Things paradigm for intelligent analysis of spa treatments to manage patients correctly.

The will to travel leads humans to discover new places and enjoy new adventures. However, tourists usually need help knowing what to visit and, avoiding time issues, in which order to explore several Points of Interest (POIs). In this field, new technologies can help tourists to improve their experiences and select the visiting path according to personal preferences. Therefore, the employment of Recommender Systems allows the personalization of the experience through the appropriate POIs’ selection. Moreover, RSs’ analysis could take advantage of contextual information that suits the personalization in the specific environment where the elaboration happens, providing users with even more specific and tailored paths. This paper aims to design personalized visiting paths combining a Context-Aware Recommender System (CARSs) and a mathematical model to maximize the number of visited POIs in the available time. The proposed approach is tested through a prototype, obtaining promising results.

Cultural Heritage Buildings need to be preserved through interventions and actions that ensure accessibility and availability to present and future generations. The diffusion o f new technologies, including low-cost sensors and devices, has introduced many possibilities and strategies to monitor environments, including Historical and Cultural value buildings. Based on the Internet of Things (IoT) paradigm, modern devices and sensors can collect and manage helpful information to build a Digital Twin of the surrounding environment. In this scenario, reproducing a Digital Twin of Cultural Heritage Buildings could be crucial to monitoring, managing, and performing action aiming to protect them. One of the challenges in recent years is the automatic analysis of collected information. This paper introduces a novel approach to protecting Buildings using Heritage Building Information Modeling (HBIM), which leverages a sensor network and Deep Learning techniques to analyze sensor data. A case study related to the Archeological Park of Pompeii will be presented, where real-time collected data is managed and shared via a web-cloud-IoT platform. Then, collected data are analyzed through a Generative Adversarial Network (GAN) to prevent future issues related to Cultural Heritage Buildings, particularly humidity inside structures, which is crucial for ancient building preservation.

In the Internet era, where information and communication technologies (ICT) allow data exchange, new tools able to select the correct data are needed. In this field, Recommender Systems have a prime location. The recommendation methods take many forms based on the information exploited in order to provide rating forecasts. However, all of them can choose the correct information to support users. Indeed, these methods allow for overcoming the information overload problem. This paper introduces the theoretical bases of a novel recommendation method defined Rating Singular Value Decomposition (RSVD). RSVD is a Content-Based method that exploits the Singular Value Decomposition properties in order to calculate rating forecasts. This method aims to elaborate the users and items profile to obtain matrices related to ones obtained in Collaborative Filtering methods that exploit Singular Value Decomposition. The accuracy of RSVD is compared with the accuracy of Collaborative Filtering methods, and a study on the sparsity problem is performed. The results obtained are promising.

The world of distance education is constantly expanding, enriching itself with tools and services to increase the ability to provide training content. Due to the new technologies, the training paths take on new appealing features; however, it remains complex to suggest the appropriate training path to the right student. In this scenario, the use of Recommender Systems (RSs) could be helpful. RSs could allow recommending personalized learning paths to students in order to improve their abilities and their knowledge. In particular, among Recommender Systems, some of them consider contextual information. This paper aims to describe a new approach that suggests learning paths to users taking advantage of recommendation techniques and introducing them through multimedia content. Moreover, the proposed approach aims to provide recommendations when ratings are unknown through the knowledge of profiles of users and items. The proposed approach has been tested through students of two courses with diverse characteristics.

In March 2020, due to the pandemic linked to the COVID-19 virus spreading, all in-person teaching activities planned at the University of Salerno were suspended. This sudden suspension has forced the academic world to rethink in a short time the didactics and to modify teaching styles and approaches, especially in those courses in which laboratory activities were planned. This article will be presented the experiences made in the Computer Networks and Protocols for the Internet of Things (IoT) course. The course includes a significant laboratory component where students must put into practice what they learned during the theoretical lessons. In particular, the laboratory exercises involve using specific equipment and devices that can build digital ecosystems for the control and management of well-defined operational contexts difficult to reproduce virtually. Therefore, to give continuity to the didactic activity, it was necessary to redesign the entire exercise system preserving the original educational objectives.

Italy offers a Cultural Heritage of considerable value to be protected. In fact, the artifacts and ancient buildings are affected by a natural deterioration linked to the flow of time. Sometimes the deterioration compromises the functionality of Cultural Heritage, driving them toward degradation. In this scenario, given the different critical points, the wide variability of the factors involved, and the wide range of possible treatments, intervene efficiently seems impossible. However, the spread of low-cost technology has led to the possibility of having various devices and sensors able to communicate and interact with each other and with humans: the Internet of Things (IoT). In this scenario, the IoT paradigm allows mapping the reality by defining a coherent virtual environment, which could help preserve Cultural Heritage. This paper aims to introduce an IoT-based system that combines three aspects: monitoring, predictive maintenance, and decision-making related to interventions to be implemented to preserve buildings belonging to Cultural Heritage. In order to test the proposed architecture, a prototype capable of interacting with expert users has been implemented and tested. The results of the experimental campaign are promising.

Summary Increasing populations and rapid large‐scale urbanization has created a demand to increase the quality of life through economic development, social stability, and better quality environments. These issues are addressed in the field of Smart Cities where, through the Internet of Things, efforts are being made to support added‐value services for the administration of the city and for citizens. The continuous exchange of information inevitably produces a huge amount of data, which demands analyses of data using unconventional methods within a Big Data context. How can we properly process these data? How can we properly use these data in order to increase the competitiveness and efficiency of services, and how could they contribute to social development? Services that could be useful in this field include Early Warning Systems. Information management environments, or more generally pervasive data contexts, may be supported by context representation approaches and enhanced through adopting probabilistic approaches such as Context Dimension Tree, Ontology, and Bayesian Network. The aim of this work is to introduce and explain a methodology for merging CDTs and Ontologies, and probabilistic approach based on BNs in order to help expert users handle emergencies and provide suggestions for improving the liveability of cities for their inhabitants.

Finding the information and making it available today is a very complex challenge. On the one hand, the large amount of available data requires a great ability to manage information; on the other hand, understanding the real needs of users requires complex systems that can provide contextual information. Italy's economy is based on tourism, thanks to its cultural heritage visited by millions of people every day. The goal of this paper is to create a recommender system to furnish a tailor-made story for the user based on the context in which it is located, thanks to a smart app that can provide contextual information. Using a Chatbot, this system also provides a context awareness data in order to enhance the tourist experience.

Today, the existing technologies, such as smartphone and other pervasive devices, can be used for context awareness and help people to undertake conscious choices.In fact, unlike what happened in the past, all of these data are managed and contextualized for the particular application.We need to understand, select, treat and finally opportunely expose these data.In this scenario, there are many Context Aware applications.This paper introduces a tailor made system for providing personalized services and it is based on a graphical formalism for the context representation: the Context Dimension Tree.This system provides the needed information about places that are of great interest for the visitors, selecting them using user preferences.A case study is applied to an event in Salerno, an Italian town, called Artist's Lights.Finally, an experimental campaign has been conducted, obtaining interesting results.

FIGURE 8. Chone kroyerii. (A) Anterior end, ventral view; (B) same, dorsal view; (C) spermatozoon; (D) body, lateral view; (E–G) paleate chaetae; (H) bayonet chaeta; (I) very long, posterior abdominal chaeta; (J) thoracic uncinus; (K) anterior abdominal uncinus; (L) posterior abdominal uncinus. (A–B, D) Methyl green staining. (A–L) [ZMO Boūgestrommen].

This paper introduces an Adaptive Context Aware Recommender system based on the Slow Intelligence approach.The system is made available to the user as an adaptive mobile application, which allows a high degree of customization in recommending services and resources according to his/her current position and global profile.A case study applied to the town of Pittsburgh has been analyzed considering various users (with different profiles as visitors, students, professors) and an experimental campaign has been conducted obtaining interesting results.

In this article, the use of a new term extraction method for query expansion (QE) in text retrieval is investigated. The new method expands the initial query with a structured representation made of weighted word pairs (WWP) extracted from a set of training documents (relevance feedback). Standard text retrieval systems can handle a WWP structure through custom Boolean weighted models. We experimented with both the explicit and pseudorelevance feedback schemas and compared the proposed term extraction method with others in the literature, such as KLD and RM3. Evaluations have been conducted on a number of test collections (Text REtrivel Conference [TREC]‐6, ‐7, ‐8, ‐9, and ‐10). Results demonstrated that the QE method based on this new structure outperforms the baseline.

The learning of a Bayesian network structure, especially in the case of wide domains, can be a complex, time-consuming and imprecise process. Therefore, the interest of the scientific community in learning Bayesian network structure from data is increasing: many techniques or disciplines such as data mining, text categorization, and ontology building, can take advantage from this process. In the literature, there are many structural learning algorithms but none of them provides good results for each dataset. This paper introduces a method for structural learning of Bayesian networks based on a MultiExpert approach. The proposed method combines five structural learning algorithms according to a majority vote combining rule for maximizing their effectiveness and, more generally, the results obtained by using of a single algorithm. This paper shows an experimental validation of the proposed algorithm on standard datasets.

Prior to 2003, mankind generated a total of about 5 Exabyte's of contents. Now, we generate this amount of contents in about two days! The spread of generic (as Twitter, Facebook or Google+) or specialized (as Linked In or Viadeo) social networks allows sharing opinions on different aspects of life every day. Therefore this information is a rich source of data for opinion mining and sentiment analysis. This paper introduces a novel approach to the sentiment analysis based on the Weighted Word Pairs obtained by the use of the Latent Dirichlet Allocation (LDA) approach. The proposed methodology aims at identifying a word-based graphical model for depicting and mining a positive or negative attitude towards a topic. For the evaluation of the proposed approach a challenging scenario has been set: the real-time analysis of tweets. The experimental evaluation shows how the proposed approach is effective and satisfactory.

It is well known that supervised text classification methods need to learn from many labeled examples to achieve a high accuracy. However, in a real context, sufficient labeled examples are not always available. For this reason, there has been recent interest in methods that are capable of obtaining a high accuracy even if the size of the training set is not big. The main purpose of text mining techniques is to identify common patterns through the observation of vectors of features and then to use such patterns to make predictions. Most existing methods usually make use of a vector of features made up of weighted words that unfortunately are insufficiently discriminative when the number of features is much higher than the number of labeled examples. In this paper we demonstrate that, to obtain a greater accuracy in the analysis and revelation of common patterns, we could employ more complex features than simple weighted words. The proposed vector of features considers a hierarchical structure, named a mixed Graph of Terms, composed of a directed and an undirected sub-graph of words, that can be automatically constructed from a set of documents through the probabilistic Topic Model. The method has been tested on the top 10 classes of the ModApte split from the Reuters-21578 dataset, learned on several subsets of the original training set and showing a better performance than a method using a list of weighted words as a vector of features and linear support vector machines.

It is well known that one way to improve the accuracy of a text retrieval system is to expand the original query with additional knowledge coded through topic-related terms. In the case of an interactive environment, the expansion, which is usually represented as a list of words, is extracted from documents whose relevance is known thanks to the feedback of the user. In this paper we argue that the accuracy of a text retrieval system can be improved if we employ a query expansion method based on a mixed Graph of Terms representation instead of a method based on a simple list of words. The graph, that is composed of a directed and an undirected subgraph, can be automatically extracted from a small set of only relevant documents (namely the user feedback) using a method for term extraction based on the probabilistic Topic Model. The evaluation of the proposed method has been carried out by performing a comparison with two less complex structures: one represented as a set of pairs of words and another that is a simple list of words.

Text classification methods have been evaluated on supervised classification tasks of large datasets showing high accuracy. Nevertheless, due to the fact that these classifiers, to obtain a good performance on a test set, need to learn from many examples, some difficulties may be found when they are employed in real contexts. In fact, most users of a practical system do not want to carry out labeling tasks for a long time only to obtain a better level of accuracy. They obviously prefer algorithms that have high accuracy, but do not require a large amount of manual labeling tasks. In this paper we propose a new supervised method for single-label text classification, based on a mixed Graph of Terms, that is capable of achieving a good performance, in term of accuracy, when the size of the training set is 1% of the original. The mixed Graph of Terms can be automatically extracted from a set of documents following a kind of term clustering technique weighted by the probabilistic topic model. The method has been tested on the top 10 classes of the ModApte split from the Reuters-21578 dataset and learned on 1% of the original training set. Results have confirmed the discriminative property of the graph and have confirmed that the proposed method is comparable with existing methods learned on the whole training set.

We present a mathematical model of multimedia data streams and a framework for multimedia functional dependency analysis. The dual objectives are to effectively design multimedia data streams schema and to efficiently process continuous queries on sensor-based multimedia data streams. To further improve query processing, we introduce the concept of ontological filtering. A software tool to add multimedia functional dependencies to ontology is developed. Based upon multimedia functional dependency analysis and ontological filtering, query processing algorithms, illustrative examples and experimental results for sensor-based multimedia data streams continuous querying are presented to demonstrate the practical applications of our approach.

Mass customization is one of the most interesting and promising approach in the e-business field. In today’s competitive global market the understanding of customer needs and desires is becoming the essential preliminary remark for a successful design and implementation of products. Product development based on customer preferences with applications of innovative technologies is an essential key in order to obtain a larger market share and faster sales growth. In this scenario a tool as the product configurator is becoming a real answer to one of most important question: how to organize product design to satisfy individual customer need without trading off cost-efficiency of mass production? This paper discusses a novel approach for the design of a smart product configurator. At this moment, in fact, the configurator is just a product viewer for the customer and does not implement any reasoning logics or user adaptive approach. So an ontology based approach is presented. In this methodology three ontologies are introduced: the customer needs ontology, the product functionalities ontology and the product configuration ontology. These ontologies represent the requirement and configuration knowledge that needs for a real customization of the product. The customer has to express his product demands by the use of natural language and by the mapping among the introduced ontologies and the use of a Bayesian Network approach the automatic conversion between customer needs and product configuration is achieved.

In this paper we introduce a tutoring approach for e-learning formative process. This approach is strictly related to the assessment phase. Assessment in the context of education is the process of characterizing what a student knows. The reasons to perform evaluation are quite varied, ranging from a need to informally understand student learning progress in a course to a need to characterize student expertise in a subject. Finding an appropriate assessment tool is a central challenge in designing a tutoring approach. In this paper we propose a method based on the use of ontologies and their representation through a Bayesian networks. The aim of our approach is the generation of adapted questionnaires in order to test the student's knowledge of every domain's subject. Analyzing the results of the evaluation an intelligent tutoring system can help students offering an effective support to learning process and adapting their learning paths

In this paper we face the problem of partitioning the news videos into stories, and of their classification according to a predefined set of categories. In particular, we propose to employ a multi-level probabilistic framework based on the hidden Markov models and the Bayesian networks paradigms for the segmentation and the classification phases, respectively. The whole analysis is carried out exploiting information extracted from the video and the audio tracks using techniques of superimposed text recognition, speaker identification, speech transcription, anchor detection. The system was tested on a database of Italian news videos and the results are very promising