Pubblicazioni

1. Lessons learned from longitudinal modeling of mobile-equipped visitors in a complex museum

Autori: Francesco Piccialli, Yuji Yoshimura, Paolo Benedusi, Carlo Ratti & Salvatore Cuomo 

Abstract: 

Cultural sites are evolving toward smart environments, including the notion of hyper-connected museums. In this context, stakeholders of cultural assets need more advanced and comprehensive ICT systems for monitoring and modeling visitorsbehaviors.

In this paper, we discuss the results of a longitudinal research study embracing multiple seasons, in a complex cultural structure including outdoor and indoor attractions, multiple floors and multiple routes for visitors. Here, interactive mobile devices were used for both offering multimedial context-aware assistance to visitors and monitoring services to museum stakeholders.

We deeply analyzed the data using an unsupervised classification approach, obtaining a model where the number of distinct user profiles and the number of features were considered not trivial as well as not too complex for museum stakeholders. We observed that some optimistic expectations about visitor performances were only partially met, devising possible explanations in terms of the different user profiles and features of the model. Finally, we also compared some outcomes from our interactive system with those obtained in another complex cultural structure using a non invasive monitoring system.

 

2. Titolo: A Deep Learning approach for Path Prediction in a Location-based IoT system

Autori: Francesco Piccialli, Fabio Giampaolo,Gianpaolo Casolla, Vincenzo Schiano Di Cola, Kenli Li 

Abstract: 

Knowing in real-time the position of objects and people, both in indoor and outdoor spaces, allows companies and organizations to improve their processes and offer new kind of services.

Nowadays Location-based Services (LBS) generate a significant amount of data thank to the widespread of the Internet of Things; since they have been quickly perceived as a potential source of profit, several companies have started to design and develop a wide range of such services. One of the most challenging research tasks is undoubtedly represented by the analysis of LBS data through Machine Learning algorithms and methodologies in order to infer new knowledge and build-up even more customized services.

Cultural Heritage is a domain that can benefit from such studies since it is characterized by a strong interaction between people, cultural items and spaces. Data gathered in a museum on visitor movements and behaviours can constitute the knowledge base to realize an advanced monitoring system able to offer museum stakeholders a complete and real-time snapshot of the museum locations occupancy. Furthermore, exploiting such data through Deep Learning methodologies can lead to the development of a predictive monitoring system able to suggest stakeholders the museum locations occupancy not only in real-time but also in the next future, opening new scenarios in the management of a museum.

In this paper, we present and discuss a Deep Learning methodology applied to data coming from a non-invasive Bluetooth IoT monitoring system deployed inside a cultural space. Through the analysis of visitors’ paths, the main goal is to predict the occupancy of the available rooms. Experimental results on real data demonstrate the feasibility of the proposed approach; it can represent a useful instrument, in the hands of the museum management, to enhance the quality-of-service within this kind of spaces.

  • Partner coinvolti: Università Federico II, Napoli

 

3. Path prediction in IoT systems through Markov Chain algorithm

Autori: Francesco Piccialli, Salvatore Cuomo, Fabio Giampaolo,Gianpaolo Casolla, Vincenzo Schiano Di Cola

Abstract: 

In the Data Technology Era, inferring knowledge from data is an ubiquitous and pervasive research topic.

Digital Ecosystems based on the Internet of Things (IoT) are generally designed for generating and collecting complex, real-time and (un)structured data. As one of the main component of the Smart City framework, the huge amount of IoT data has to be opportunely processed, also through Machine Learning algorithms in order to discover new knowledge and to improve the quality-of-life of the citizens.

In our research work we propose some learning methodologies to analyse and forecast visitors’ paths within a cultural and complex space. Starting from data collected in a museum equipped with a non-invasive monitoring IoT system, we show how it is possible to discover and predict useful information on visitors’ movements and, finally, we present and discuss some useful insights on their behaviours within a real case-of-study.

 

4. Path Unsupervised learning on multimedia data: a Cultural Heritage case study 

Autori: Francesco Piccialli,Gianpaolo Casolla, Salvatore Cuomo, Fabio Giampaolo, Edoardo Prezioso, Vincenzo Schiano Di Cola

Abstract: 

Integrating and analyzing a large amount of data extracted from different sources can be considered a key asset for businesses, organizations, research institutions that also deal with the Cultural Heritage domain. In the last decade, Internet of Things (IoT) technologies and the massive use of mobile devices contributed to generate an enormous flow of multimedia data, whose collection, analysis and interpretation allows for real-time analysis related to the behaviours, preferences and opinions of users. In this paper we present and discuss an unsupervised learning approach on multimedia features of a dataset coming from an Internet of Things framework.

The main research objective of this work is to assess how the collection of behavioural IoT data coming from the Cultural Heritage domain can be opportunely exploited by means of unsupervised learning techniques in order to produce useful insights for the stakeholders, especially considering the multimedia features of such data. The presented experimental results, executed in a real case study, assess how the Cultural Heritage domain, and the related stakeholders, can benefit from these kind of services and applications.

 

5. An IoT data analytics approach for cultural heritage

Autori: Francesco Piccialli, Paolo Benedusi,Luca Carratore, Giovanni Colecchia

Abstract: 

 The ability to integrate, manage, and analyze large amounts of data extracted from different sources is becoming a key asset for businesses, organizations, and research institutions that deal with the cultural heritage domain. Nowadays, it is well known that modern technologies and the massive use of mobile devices can contribute to generate an enormous flow of data, whose collection, analysis, and interpretation allows for real-time analysis related to the behaviors, preferences, and opinions of users.

In this paper, we present and discuss a data analytics approach relying on an Internet of Things framework. The main goal is to assess how the collection of behavioral IoT data coming from the cultural heritage domain can be opportunely exploited by means of data science and data analytics techniques in order to produce useful insights.

Experimental results performed in a real case study demonstrate how the cultural heritage domain, and the related stakeholders, can benefit from these kind of applications.

 

6. Decision making in IoT environment through unsupervised learning

Autori: Francesco Piccialli,Gianpaolo Casolla, Salvatore Cuomo, Fabio Giampaolo,Vincenzo Schiano Di Cola

Abstract:

Nowadays, unsupervised learning can provide new perspectives to identify hidden patterns and classes inside the huge amount of data coming from the Internet of Things (IoT) world.

Analyzing IoT data through machine learning techniques requires the use of mathematical algorithms, computational techniques, and an accurate tuning of the input parameters.

In this article, we present a study of unsupervised learning techniques applied on IoT data to support decision-making processes inside intelligent environments. To assess the proposed approach, we discuss two case studies in which behavioral IoT data have been collected, also in a noninvasive way, in order to achieve an unsupervised classification that can be adopted during a decision-making process.

The use of unsupervised learning techniques is acquiring a key role to complement the more traditional services with new decision-making ones supporting the needs of companies, stakeholders, and consumers.

  • Partner coinvolti: Università Federico II, Napoli

 

7. Exploring unsupervised learning techniques for the Internet of Things

Autori: Gianpaolo Casolla, Salvatore Cuomo, Vincenzo Schiano Di Cola, Francesco Piccialli

Abstract:

Nowadays, machine learning (ML) techniques can provide new perspectives to identify hidden patterns and classes inside data.

Applying ML to the Internet of Things (IoT) and its produced data represents a great challenge in every application domain, since analyzing IoT data increasingly requires the use of advanced mathematical algorithms, novel computational techniques, and services.

In this article, we present and discuss the application of unsupervised learning techniques on IoT data collected in a cultural heritage framework. Behavioral data have been gathered in a noninvasive way in order to achieve an ML classification that can be exploited by cultural stakeholders in terms of the medium- to long-term strategy and also in terms of strictly operational decisions. The application of ML and other learning techniques will acquire a key role to complement the more traditional services with new intelligent ones able to satisfy the needs of companies, stakeholders, and consumers.

  • Partner coinvolti: Università Federico II, Napoli

 

8.Cultural Control Room: CETRA addresses and exploits the IoT to innovate Cultural Heritage Promotion 

Autori: Sam Habibi Minelli, Paolo Ongaro, Daniele Ugoletti, Rubino Saccoccio and Simona Maresca

Abstract:

Abstract: soluzioni tecnologiche rivolte al patrimonio culturale  e alla user experience. Una cabina di regia che consente ai manager culturali di pubblicare guide e informazioni sugli apparati di musei, archivi ed organizzazioni culturali.