Search

PurposeThe recent innovations of Industry 4.0 have made it possible to easily collect data related to a production environment. In this context, information about industrial equipment – gathered by proper sensors – can be profitably used for supporting predictive maintenance (PdM) through the application of data-driven analytics based on artificial intelligence (AI) techniques. Although deep learning (DL) approaches have proven to be a quite effective solutions to the problem, one of the open research challenges remains – the design of PdM methods that are computationally efficient, and most importantly, applicable in real-world internet of things (IoT) scenarios, where they are required to be executable directly on the limited devices’ hardware.Design/methodology/approachIn this paper, the authors propose a DL approach for PdM task, which is based on a particular and very efficient architecture. The major novelty behind the proposed framework is to leverage a multi-head attention (MHA) mechanism to obtain both high results in terms of remaining useful life (RUL) estimation and low memory model storage requirements, providing the basis for a possible implementation directly on the equipment hardware.FindingsThe achieved experimental results on the NASA dataset show how the authors’ approach outperforms in terms of effectiveness and efficiency the majority of the most diffused state-of-the-art techniques.Research limitations/implicationsA comparison of the spatial and temporal complexity with a typical long-short term memory (LSTM) model and the state-of-the-art approaches was also done on the NASA dataset. Despite the authors’ approach achieving similar effectiveness results with respect to other approaches, it has a significantly smaller number of parameters, a smaller storage volume and lower training time.Practical implicationsThe proposed approach aims to find a compromise between effectiveness and efficiency, which is crucial in the industrial domain in which it is important to maximize the link between performance attained and resources allocated. The overall accuracy performances are also on par with the finest methods described in the literature.Originality/valueThe proposed approach allows satisfying the requirements of modern embedded AI applications (reliability, low power consumption, etc.), finding a compromise between efficiency and effectiveness.

Artificial intelligence (AI) has advanced rapidly, but it has limited impact on biomedical text understanding due to a lack of annotated datasets (a.k.a. few-shot learning). Multi-task learning, which uses data from multiple datasets and tasks with related syntax and semantics, has potential to address this issue. However, the effectiveness of this approach heavily relies on the quality of the available data and its transferability between tasks. In this paper, we propose a framework, built upon a state-of-the-art multi-task method (i.e. MT-DNN), that leverages different publicly available biomedical datasets to enhance relation extraction performance. Our model employs a transformer-based architecture with shared encoding layers across multiple tasks, and task-specific classification layers to generate task-specific representations. To further improve performance, we utilize a knowledge distillation technique. In our experiments, we assess the impact of incorporating biomedical datasets in a multi-task learning setting and demonstrate that it consistently outperforms state-of-the-art few-shot learning methods in cases of limited data. This results in significant improvement across most datasets and few-shot scenarios, particularly in terms of recall scores.

Hard disk drive failures are one of the most common causes of service downtime in data centers. Predictive maintenance techniques have been adopted to extend the Remaining Useful Life (RUL) of these drives, and minimize service shortage and data loss. Several approaches based on machine and deep learning techniques have been proposed to address these issues, mostly exploiting models based on Self-Monitoring analysis and Reporting Technology (SMART) attributes. While these models have proven to be reliable, their performance is affected by the lack of information about the proximity of disk failure in time. Moreover, many of these techniques are sensitive to the highly unbalanced nature of existing data-sets, in terms of good to failed hard disk ratio. In this paper, we propose a LSTM based model combining SMART attributes and temporal analysis for estimating a hard drive health status according to its time to failure. Our approach outperforms state-of-the-art methods when evaluated on two data-sets, one containing hourly samples from 23,395 disks and the other reporting daily samples from 29,878 disks. Experimental results showed that our approach is well suited to data-sets with different sampling periods, being able to predict hard drive health status up to 45 days before failure.

Social Networks represent a cornerstone of our daily life, where the so-called social reviewing systems (SRSs) play a key role in our daily lives and are used to access data typically in the form of reviews. Due to their importance, social networks must be trustworthy and secure, so that their shared information can be used by the people without any concerns, and must be protected against possible attacks and misuses. One of the most critical attacks against the reputation system is represented by mendacious reviews. As this kind of attacks can be conducted by legitimate users of the network, a particularly powerful solution is to exploit trust management, by assigning a trust degree to users, so that people can weigh the gathered data based on such trust degrees. Trust management within the context of SRSs is particularly challenging, as determining incorrect behaviors is subjective and hard to be fully automatized. Several attempts in the current literature have been proposed; however, such an issue is still far from been completely resolved. In this study, we propose a solution against mendacious reviews that combines fuzzy logic and the theory of evidence by modeling trust management as a multicriteria multiexpert decision making and exploiting the novel concept of time-dependent and content-dependent crown consensus. We empirically proved that our approach outperforms the main related works approaches, also in dealing with sockpuppet attacks.

During the last decade, more and more Artificial Intelligence systems have been designed using complex and sophisticated architectures to reach unprecedented predictive performance. The side effect is an increase in opacity of their inner workings which is inadmissible when such systems are applied in critical domains (healthcare, finance and so on). The eXplainable AI (XAI) research field aims to overcome this limitation thus helping humans to understand black-box decisions. In this paper we propose a novel model-agnostic XAI technique, named Pivot-Aided Space Transformation for Local Explanations (PASTLE), which exploits an instance-space transformation to explain any model’s predictions, aiming to enhance human trust towards the AI decisions. We experimentally evaluate the effects of the introduced space transformation on various real-world data sets and our user study reveals promising results in terms of effective explainability.

Today, major corporations and government organizations must face the reality that they will be hacked by malicious actors. In this paper, we consider the case of defending enterprises that have been successfully hacked by imposing additional a posteriori costs on the attacker. Our idea is simple: for every real document $d$ d , we develop methods to automatically generate a set $Fake(d)$ F a k e ( d ) of fake documents that are very similar to $d$ d . The attacker who steals documents must wade through a large number of documents in detail in order to separate the real one from the fakes. Our $\mathsf {FORGE}$ FORGE system focuses on technical documents (e.g., engineering/design documents) and involves three major innovations. First, we represent the semantic content of documents via multi-layer graphs (MLGs). Second, we propose a novel concept of “meta-centrality” for multi-layer graphs. A meta-centrality (MC) measure takes a classical centrality measure (for ordinary graphs, not MLGs) as input, and generalizes it to MLGs. The idea is to generate fake documents by replacing concepts on the basis of meta-centrality with related concepts according to an ontology. Our third innovation is to show that the problem of generating the set $Fake(d)$ F a k e ( d ) of fakes can be viewed as an optimization problem. We prove that this problem is NP-complete and then develop efficient heuristics to solve it in practice. We ran detailed experiments on two datasets: one a panel of 20 human subjects, another with a panel of 10. Our results show that $\mathsf {FORGE}$ FORGE generates highly believable fakes.

Predictive Maintenance has become an important component in modern industrial scenarios, as a way to minimize down-times and fault rate for different equipment. In this sense, while machine learning and deep learning approaches are promising due to their accurate predictive abilities, their data-heavy requirements make them significantly limited in real world applications. Since one of the main issues to overcome is lack of consistent training data, recent work has explored the possibility of adapting well-known deep-learning models for image recognition, by exploiting techniques to encode time series as images. In this paper, we propose a framework for evaluating some of the best known time series encoding techniques, together with Convolutional Neural Network-based image classifiers applied to predictive maintenance tasks. We conduct an extensive empirical evaluation of these approaches for the failure prediction task on two real-world datasets (PAKDD2020 Alibaba AI OPS Competition and NASA bearings), also comparing their performances with respect to the state-of-the-art approaches. We further discuss advantages and limitation of the exploited models when coupled with proper data augmentation techniques.

Nowadays, recruitment processes are increasingly being automated by intelligent systems which provide best candidates for companies’ open positions, and vice versa. However, extracting information from the unstructured documents involved in these processes (e.g. resumes, jobs’ descriptions) still represents an open challenge because of their high heterogeneity (in the form and style) and the lack of pre-defined standards between different companies and/or countries. In this paper, we address the resume information extraction problem, focusing on documents within the Italian Labor Market. Specifically, we propose an effective and efficient end-to-end framework capable of providing a complete candidate overview including his personal information, skills and work experiences. Specifically, after having extracted the raw data from the resume documents, the system segments them into semantically consistent parts using linguistics patterns. Each segment is further processed with a NER algorithm, based on pre-trained language models, to extract relevant information which an HR specialist could consult in order to assess the suitability of a candidate for a job offer. We collected (and labeled) a new Italian resume dataset and our results prove the effectiveness of the proposed method, especially considering the great advantages our segmentation strategy brings to the NER performance with respect to standard line-based segmentation approaches. In addition, our system achieves promising performance when combined with modern NLP models.

In this paper we describe a novel algorithm based on Game Theory for Community Detection in Online Social Networks. Extending several Game Theoretic approaches that are well established in the literature, we propose a novel algorithm that outperforms the previous ones in terms of computational complexity and efficiency of results, as proved by a number of experiments on standard data sets.

Recommender systems are increasingly playing an important role in our life, enabling users to find “what they need” within large data collections and supporting a variety of applications, from e-commerce to e-tourism. In this paper, we present a Big Data architecture supporting typical cultural heritage applications. On the top of querying, browsing, and analyzing cultural contents coming from distributed and heterogeneous repositories, we propose a novel user-centered recommendation strategy for cultural items suggestion. Despite centralizing the processing operations within the cloud, the vision of edge intelligence has been exploited by having a mobile app ( Smart Search Museum ) to perform semantic searches and machine-learning-based inference so as to be capable of suggesting museums, together with other items of interest, to users when they are visiting a city, exploiting jointly recommendation techniques and edge artificial intelligence facilities. Experimental results on accuracy and user satisfaction show the goodness of the proposed application.

When considering popular On-line Social Networks (OSN) containing heterogeneous multimedia data sources, the complexity of the underlying processing systems becomes challenging, and requires to implement application-specific but still comprehensive benchmarking. The variety of big data architectures (and of their possible realization) for both batch and streaming processing in a huge number of application domains, makes the benchmarking of these systems critical for both academic and industrial communities. In this work, we evaluate the performance of two state-of-art big data architectures, namely Lambda and Kappa, considering OSN data analysis as reference task. In more details, we have implemented and deployed an influence analysis algorithm on the Microsoft Azure public cloud platform to investigate the impact of a number of factors on the performance obtained by cloud users. These factors comprise the type of the implemented architecture, the volume of the data to analyze, the size of the cluster of nodes realizing the architectures and their characteristics, the deployment costs, as well as the quality of the output when the analysis is subjected to strict temporal deadlines. Experimental campaigns have been carried out on the Yahoo Flickr Creative Commons 100 Million (YFCC100M). Reported results and discussions show that Lambda outperforms Kappa architecture for the class of problems investigated. Providing a variety of analyses – e.g., also investigating the impact of dataset size, scaling, cost – this paper provides useful insights on the performance of these state-of-art big data architectures that are helpful to both experts and newcomers interested in deploying big data architectures leveraging cloud platforms.

With Artificial Intelligence becoming part of a rapidly increasing number of industrial applications, more and more requirements about their transparency and trustworthiness are being demanded to AI systems, especially in military, medical and financial domains, where decisions have a huge impact on lives. In this paper, we propose a novel model-agnostic Explainable AI (XAI) technique, named Cluster-aided Space Transformation for Local Explanation (CASTLE), able to provide rule-based explanations based on both the local and global model’s workings, i.e. its detailed ”knowledge” in the neighborhood of the target instance and its general knowledge on the training dataset, respectively. The framework has been evaluated on six datasets in terms of temporal efficiency, cluster quality and model significance. Eventually, we asked 36 users to evaluate the explainability of the framework, getting as result an increase of interpretability of 6 % with respect to another state-of-the-art technique, named Anchors.

The exponential growth of connected devices (i.e. laptops, smartphones or tablets) has radically changed communications means, also making it faster and impersonal by using On-line Social Networks and Instant messaging through several apps. In this paper we discuss about the cyberbullying problem, focusing on the analysis of the state-of-the-art approaches that can be classified in four different tasks (Binary Classification, Role Identification, Severity Score Computation and Incident prediction). In particular, the first task aims to predict if a particular action is aggressive or not based on the analysis of different features. In turn, the second and the third task investigate the cyberbullying problem by identifying users’ role in the exchanged message or assigning a severity score to a given users or session respectively. Nevertheless, information heterogeneity, due to different multimedia contents (i.e. text, emojis, stickers or gifs), and the use of datasets, which are typically unlabeled or manually labelled, create continuous challenges in addressing the cyberbullying problem.

In this paper we challenge the issue of detecting anomalous events in computer systems log files, through a novel graph mining approach. The basic idea is to model log temporal sequences as a particular graph and event detection as a particular path finding problem. Thus, anomalous sequences correspond to log parts that can not be “explained” by any path in the graph. We propose a novel Iterative Partitioning Log Mining technique to parse any kind of logs and to model their temporal sequence as a probabilistic penalty graph. The approach has been implemented in a framework supporting both real time and batch processing realized on the top of the Apache Spark analytics engine for large-scale data processing. Experimental results show the advantages of the proposed framework in terms of effectiveness for different system configurations.

Social media recommendation differs from traditional recommendation approaches as it needs considering not only the content information and users' similarities, but also users' social relationships and behavior within an online social network as well. In this article, a recommender system – designed for big data applications – is used for providing useful recommendations in online social networks. The proposed technique represents a collaborative and user-centered approach that exploits the interactions among users and generated multimedia contents in one or more social networks in a novel and effective way. The experiments performed on data collected from several online social networks show the feasibility of the approach towards the social media recommendation problem.

One of the most important challenge in the Industry 4.0 era is to provide advanced maintenance mechanisms able to avoid or minimize failures during production processes. In this paper, we propose a novel approach to the predictive maintenance problem that combines encoding strategies to convert temporal data into images and classical deep learning strategies. In particular, the basic idea is to exploit GAF (Gramian Angular Field) encoding to obtain images from time series related to production systems, which can be used in pre-trained convolutionary neural networks for better prediction performance and thus to create a more efficient and simple predictive maintenance scheme. Some experiments on several standard datasets show the advantages of the proposed techniques w.r.t. to other approaches in the literature.

Background: Neonatal infections represent one of the six main types of healthcare-associated infections and have resulted in increasing mortality rates in recent years due to preterm births or problems arising from childbirth. Although advances in obstetrics and technologies have minimized the number of deaths related to birth, different challenges have emerged in identifying the main factors affecting mortality and morbidity. Dataset characterization: We investigated healthcare-associated infections in a cohort of 1203 patients at the level III Neonatal Intensive Care Unit (ICU) of the “Federico II” University Hospital in Naples from 2016 to 2020 (60 months). Methods: The present paper used statistical analyses and logistic regression to identify an association between healthcare-associated blood stream infection (HABSIs) and the available risk factors in neonates and prevent their spread. We designed a supervised approach to predict whether a patient suffered from HABSI using seven different artificial intelligence models. Results: We analyzed a cohort of 1203 patients and found that birthweight and central line catheterization days were the most important predictors of suffering from HABSI. Conclusions: Our statistical analyses showed that birthweight and central line catheterization days were significant predictors of suffering from HABSI. Patients suffering from HABSI had lower gestational age and birthweight, which led to longer hospitalization and umbilical and central line catheterization days than non-HABSI neonates. The predictive analysis achieved the highest Area Under Curve (AUC), accuracy and F1-macro score in the prediction of HABSIs using Logistic Regression (LR) and Multi-layer Perceptron (MLP) models, which better resolved the imbalanced dataset (65 infected and 1038 healthy).

Researchers and companies are making great efforts to create new ways of interaction with a increasing number and types of electronic devices, with particular attention to the conversational interface, whether spoken or written. The cultural heritage domain can bring many benefits from the great effort in this field, as a smart conversational system can provide both room/hotel/apartment information and information related to the area, natural or cultural points of interest, tourist routes, social events, history etc. The paper’s main goal is to present a quick and effective service to the tourists visiting a city for the first time. Learning the cultural preferences of the users can enhance customer satisfaction, since a smart system can propose them customised tours in order to reach point of interests according to the expressed preferences. The interaction with the user is simplified by a conversational interface (chatbot). © Springer Nature Switzerland AG 2021.

Failure rates of Hard Disk Drives (HDDs) are high and often due to a variety of different conditions. Thus, there is increasing demand for technologies dedicated to anticipating possible causes of failure, so to allow for preventive maintenance operations. In this paper, we propose a framework to predict HDD health status according to a long short-term memory (LSTM) model. We also employ eXplainable Artificial Intelligence (XAI) tools, to provide effective explanations of the model decisions, thus making the final results more useful to human decision-making processes. We extensively evaluate our approach on standard data-sets, proving its feasibility for real world applications.

The technology improvement has radically changed how tourists perform their journey offering new services to enhance their cultural experience and to easily retrieve required information. In this paper we propose a novel framework, that models intangible and tangible cultural objects into a unified data model, for supporting tourists journey. In particular, a Micro-service architecture has been designed to provide several services whose tourists can access through a conversational agent based on the Seq2Seq model. Furthermore, we propose also an Enterprise Service Bus to ingest events automatically from promotional website or manually from public and/or private organization. We have evaluated the proposed framework according to the following two experiments: (i) the efficiency and efficacy of the Chatbot engine, showing that the use of GRU cells allows to obtain better results in terms of loss and accuracy with respect to LSTM one and (ii) the efficacy of the proposed framework according to the NASA-TLX asking to 50 users to interact with the Chatbot, demonstrating that the proposed approach outperforms the state-of-the art ones combining natural processing language, user profile information and historical activities.

In modern Online Social Networks (OSNs), the need to detect users’ communities based on their interests and social connections has became a more and more important challenge in literature. Community Detection supports and make more effective and efficient several Social Network Analysis (SNA) applications: the diffusion of a new idea or technologies can be maximized by identifying of people group interested about a given topic, the recommendation suggestion can be improved taking in account also how the social ties can be influenced the user chooses and the behaviors of people in the same communities, expert finding tasks could be more accurate if users are earlier subdivided into thematic groups, and so on. This paper presents a survey that provides a comprehensive and comparative study of all the different community detection techniques applicable to the various models proposed for OSNs. In particular, the most diffused approaches based on game theory, artificial intelligence and fuzzy strategies are detailed and compared, highlighting the related pros and cons. In addition, the problem of their applicability on the different OSN models is discussed, focusing on complex networks. Finally, the main open issues and challenges for the community detection problem are reported to address the futures work concerning this topic.

In this paper we describe a novel algorithm based on Game Theory for Community Detection in Social Networks. Extending several Game Theoretic approaches that are well established in the literature, we propose a novel algorithm that outperforms the previous ones in terms of computational complexity and efficiency of results, as proved by a number of experiments on standard data sets. © 2019 IEEE.

Community Detection in On Line Social Networks is a classic feature in networked systems, from the fields of biology, economics, politics and computer science, as well. This paper describes a novel Community Detection method based on a deep learning approach, facing the challenging problems related to the dimensions of the involved data structures, and proposing a novel convolutional technique particularly useful for sparse matrices. Several experiments have been reported and discussed in real scenarios.

Nowadays, every facet of people’s lifestyle is impacted by the continuous use of technologies and in particular the Internet. Social interactions have been radically changed since new technologies have removed communication barriers. The obstacles of time and space have been overcome, letting people from different places and cultures communicate. People tend to become part of a dense social network, where the distribution of information becomes an almost immediate process. This is the reason why companies, public institutions and business activities have opted for social networks as communication medium. As a consequence, people seem to incorporate information gained from the social networks into their decision-making processes.

Online social network (OSN) is quickly becoming a promising field of interest for big data analytics and for a number of real applications, ranging from marketing to user profiling and recommendation. Anyway, although OSNs are naturally formed by heterogeneous data, in the recent past, only a few works have considered an explicit use of multimedia in their models. In this paper, we explicitly take into account the intrinsic characteristics of multimedia, having the awareness that in this way, both the models and the analysis algorithms will enormously benefit from such kind of information. In particular, we describe a multimedia data model for OSN, in order to provide, in a unique framework, novel mechanisms for effective management of multimedia information supporting several classic applications, such as influence analysis and maximization. Experiments have been carried out on Flickr Creative Commons Y FCC100M dataset containing about 100 million images, showing that the proposed approach combines both time and efficacy performances.

Credit risk assessment plays a key role for correctly supporting financial institutes in defining their bank policies and commercial strategies. Over the last decade, the emerging of social lending platforms has disrupted traditional services for credit risk assessment. Through these platforms, lenders and borrowers can easily interact among them without any involvement of financial institutes. In particular, they support borrowers in the fundraising process, enabling the participation of any number and size of lenders. However, the lack of lenders’ experience and missing or uncertain information about borrower’s credit history can increase risks in social lending platforms, requiring an accurate credit risk scoring. To overcome such issues, the credit risk assessment problem of financial operations is usually modeled as a binary problem on the basis of debt’s repayment and proper machine learning techniques can be consequently exploited. In this paper, we propose a benchmarking study of some of the most used credit risk scoring models to predict if a loan will be repaid in a P2P platform. We deal with a class imbalance problem and leverage several classifiers among the most used in the literature, which are based on different sampling techniques. A real social lending platform (Lending Club) data-set, composed by 877,956 samples, has been used to perform the experimental analysis considering different evaluation metrics (i.e. AUC, Sensitivity, Specificity), also comparing the obtained outcomes with respect to the state-of-the-art approaches. Finally, the three best approaches have also been evaluated in terms of their explainability by means of different eXplainable Artificial Intelligence (XAI) tools.

In this paper the authors define a novel data model for Multimedia Social Networks (MSNs), i.e. networks that combine information on users belonging to one or more social communities together with the multimedia content that is generated and used within the related environments. The proposed model relies on the hypergraph data structure to capture and to represent in a simple way all the different kinds of relationships that are typical of social networks and multimedia sharing systems, and in particular between multimedia contents, among users and multimedia content and among users themselves. Different applications (e.g. influence analysis, multimedia recommendation) can be then built on the top of the introduce data model thanks to the introduction of proper user and multimedia ranking functions. In addition, the authors provide a strategy for hypergraph learning from social data. Some preliminary experiments concerning efficiency and effectiveness of the proposed approach for analysis of Last.fm network are reported and discussed.

Social networks analysis is an emerging trend among scholars and researchers in the last years. A great number of companies are interested in social networks data mining. Data gathered from Facebook, Twitter or other social networks result to be very attractive in many application fields, like economics analysis, sentiment analysis, and politics analysis and so on. In our work, we focused on the analysis of the content of textual information obtained from the social media. Our investigation is finalized to extract hot topics in social network. We considered, as case study, reviews obtained from the social network Yelp.

Despite the great amount of research done in the Online Social Networks (OSNs) field, only few works have investigated the use of multimedia data in such realm. Instead, it is the authors’ opinion that a novel data model that takes into account the intrinsic characteristics of multimedia may be of great help in managing Multimedia OSNs for providing more effective algorithms. In this paper, we describe a novel OSN data model that supports easy management of multimedia content in a unique framework, providing a more effective and efficient mechanism for data and information management in a variety of applications, especially for Influence Analysis aims.

In this paper, we propose and describe a novel recommender system for big data applications that provides recommendations on the base of the interactions among users and generated multimedia contents in one or more social media networks, leveraging a collaborative and user-centered approach. Preliminary experiments using data of several online social networks show how our approach obtains very promising results.

In the last few years, the diffusion of the Internet of Things (IoT) architectures is at the basis of the explosive growth of data and information useful to build adaptive, intelligent, autonomous and non-intrusively systems. In this work, we explore the Cultural Heritage domain and propose an innovative methodology for adaptive and context-aware system design, presenting a semantic model of intelligent environments and rule-based application behavior.

Social Network Analysis has been introduced to study the properties of Online Social Networks for a wide range of real life applications. In this paper, we propose a novel methodology for solving the Influence Maximization problem, i.e. the problem of finding a small subset of actors in a social network that could maximize the spread of influence. In particular, we define a novel influence diffusion model that, learning recurrent user behaviours from past logs, estimates the probability that a given user can influence the other ones, basically exploiting user to content actions. A greedy maximization algorithm is then adopted to determine the final set of influentials in the network. Preliminary experimental results shows the goodness of the proposed approach, especially in terms of efficiency, and encourage future research in such direction.

In this paper, inspired by hypergraph-based approaches, we propose a novel data model for social media networks: it allows to represent in a simple way all the different kinds of relationships that are typical of these environments (among multimedia contents, among users and multimedia content and among users themselves) and to enable several kinds of analytics and applications. From the other hand, we have tested several influence maximization algorithms leveraging the introduced network structure in order to show the advantages to consider also “user-to-multimedia” relationships (in addition to the “user-to-user” ones) in the influence analysis problem. Preliminary experiments using data of several social media networks shows how our approach obtains very promising results.

Online Social Networks (OSNs) have found widespread applications in every area of our life. A large number of people have signed up to OSN for different purposes, including to meet old friends, to choose a given company, to identify expert users about a given topic, producing a large number of social connections. These aspects have led to the birth of a new generation of OSNs, called Multimedia Social Networks (MSNs), in which user-generated content plays a key role to enable interactions among users. In this work, we propose a novel expert-finding technique exploiting a hypergraph-based data model for MSNs. In particular, some user-ranking measures, obtained considering only particular useful hyperpaths, have been profitably used to evaluate the related expertness degree with respect to a given social topic. Several experiments on Last.FM have been performed to evaluate the proposed approach&rsquo, s effectiveness, encouraging future work in this direction for supporting several applications such as multimedia recommendation, influence analysis, and so on.

In this paper we present a preliminary work concerning the definition of a novel data model for Multimedia Social Networks (MSNs), i.e. networks that combine information on users belonging to one or more social communities together with the multimedia content that is generated and used within a related environment. The proposed model is based on the hypergraph structure and allows us to represent in a simple way all the different kinds of relationships that are typical of a social network, in particular between multimedia content, between users and multimedia content and between users themselves, at the same time supporting several kinds of applications by means of the introduction of several ranking functions. In addition, we provide a strategy for mapping the proposed model into an object relational data model, for efficiently storing the related data. We also describe a first prototype that can store and integrate information from different OSNs (Facebook, Twitter) and multimedia sharing systems (Flickr, Lastfm), providing some facilities for browsing the entire hypergraph.

Social media provide a great amount of valuable information in the form of messages posted by users. Information extracted from posts can be considered like features giving insights about the preferences of users towards certain events. These features can be used to generate recommendations looking forward for upcoming events they might find interesting. In this work we present system for opinion analysis from tweets and recommendation of cultural heritage events. At this aim, we detect the events of interest from Tweets and propose a methodology for associating a sentiment degree with a tweet using NLP techniques.

This paper proposes a new distributed data structure based on binary trees to support k-nearest neighbor queries over very large databases. The indexing structure is distributed across a network of “peers”, where each one hosts a part of the tree and communication among nodes is realized by message passing. The advantages of this kind of approach are mainly two: it is possible to (i) handle a larger number of nodes and points than a single peer based architecture and (ii) to manage in an efficient way computation of multiple queries. In particular, we propose a novel version of the k-nearest neighbor algorithm that is able to start the query in a randomly chosen peer. Preliminary experiments have demonstrated that in about 65 % of cases a query, which starts in random node, does not involve the peer containing the root of the tree.

In this paper, we present a framework supporting information retrieval over corpora of documents using an automatic semantic query expansion approach. The main idea is to expand the set of words used as query terms exploiting the notion of semantic similarity between the concepts related to the search terms. We leverage existing lexical resources and similarity metrics computed among terms to generate - by a proper mapping into a vectorial space - an index for the fast retrieval of a set of terms “semantically correlated” to a given query term. The vector of expanded terms is then exploited in the query stage to retrieve documents that are significantly related to specific combinations of the query terms. Preliminary experimental results concerning efficiency and effectiveness of the proposed approach are reported and discussed.

Summary Nowadays, social networks provide users an interactive platform to create and share heterogeneous content for a lot of different purposes (eg, to comment events and facts, and express and share personal opinions on specific topics), allowing millions of individuals to create online profiles and share personal information with vast networks known and sometimes also unknown people. Knowledge about users, content, and relationships in a social network may be used for an adversary attack of some victims easily. Although a number of works have been done for data privacy preservation on relational data, they cannot be applied in social networks and in general for big data analytics. In this paper, we first propose a novel data model that integrates and combines information on users belonging to 1 or more heterogeneous online social networks, together with the content that is generated, shared, and used within the related environments, using an hypergraph data structure; then we implemented the most diffused centrality measures and also introduced a new centrality measure—based on the concept of “neighborhood” among users—that may be efficiently applied for a number of data privacy issues, such as lurkers and neighborhood attack prevention, especially in “interest-based” social networks. Some experiments using the Yelp dataset are discussed.

In this paper, we present a general framework for retrieving relevant information from news papers that exploits a novel summarization algorithm based on a deep semantic analysis of texts. In particular, we extract from each Web document a set of triples (subject, predicate, object) that are then used to build a summary through an unsupervised clustering algorithm exploiting the notion of semantic similarity. Finally, we leverage the centroids of clusters to determine the most significant summary sentences using some heuristics. Several experiments are carried out using the standard DUC methodology and ROUGE software and show how the proposed method outperforms several summarizer systems in terms of recall and readability.