Giuseppe D'Aniello

A multi-agent fuzzy consensus model in a Situation Awareness framework

Graphical abstractDisplay Omitted In order to define systems enabling the automatic identification of occurring situations, numerous approaches employing intelligent software agents to analyse data coming from deployed sensors have been proposed. Thus, it is possible that more agents are committed to monitor the same phenomenon in the same environment. Redundancy of sensors and agents is needed, for instance, in real world applications in order to mitigate the risk of faults and threats. One of the possible side effects produced by redundancy is that agents, observing the same phenomenon, could provide discordant opinions. Indeed, solid mechanisms for reaching an agreement among these agents and produce a shared consensus on the same observations are needed. This paper proposes an approach to integrate a fuzzy-based consensus model into a Situation Awareness framework. The main idea is to consider intelligent agents as experts claiming their opinions (preferences) on a phenomenon of interest.

A personality based adaptive approach for information systems

We defined a new adaptive approach to suggest the best interaction to the users.We get the personality of the users by inferring it from the social networks.The adaptive system instantiates for each user the best process and interface.The proposed approach includes two different layers of personalization.The approach suggests the interaction process for collaborative learning. In every context where the objective is matching needs of the users with fitting answers, the high-level performance becomes a requirement able to allow systems being useful and effective. The personalization may affect different moments of computer-humans interaction routing the users to the best answers to their needs. The most part of this complex elaboration is strictly related with the needs themselves and the residual is independent from it. It is what we may face by getting personality traits of the users.In this paper, we describe an approach that is able to get the personality of the users by inferring it from the social activities they do in order to drive them to the interactive processes they should prefer. This may happens in a wide set of situations, when they are deepened in a collaborative learning experience, in an information retrieval problem, in an e-commerce process or in a general searching activity.We defined a complete model to realize an adaptive system that may interoperate with information systems and that is able to instantiate for all the users the processes and the interfaces able to give them the best feeling and to the system the highest possible performance.

Towards Perception-Oriented Situation Awareness Systems

This paper proposes a new approach for identifying situations from sensor data by using a perception-based mechanism that has been borrowed from humans: sensation, perception and cognition. The proposed approach is based on two phases: low-level perception and high-level perception. The first one is realized by means of semantic technologies and allows to generate more abstract information from raw sensor data by also considering knowledge about the environment. The second one is realized by means of Fuzzy Formal Concept Analysis and allows to organize and classify abstract information, coming from the first phase, by generating a knowledge representation structure, namely lattice, that can be traversed to obtain information about occurring situation and augment human perception. The work proposes also a sample scenario executed in the context of an early experimentation.

Self-regulated learning with approximate reasoning and situation awareness

We present a decision support system for seamless and self-regulating learning. The decision support system presents a degree of novelty in supporting learners since it allows to: (1) understand the concepts that a learner may have acquired during her/his daily life activities, and (2) make the learner aware of these concepts and enforcing learning paths. Two key ideas are behind our results. The first idea relates to the identification of classes of indiscernible competences, and comes from the intuition that some real-world activities can lead to the acquisition of sets of competences (not always easy to discriminate) which can be considered as good approximations of competences related to a specific concept. Classes of indiscernible competences are building blocks that our decision support system uses to understand, with a certain degree of approximation, the concepts that a learner may have acquired, and this is an added value with regard to the self-awareness of a learner. The second idea is to allow our decision support system to identify incremental learning situations, which are situations in which a learning path is enforced or modified by the recognition that some concepts may have been learned, also during the execution of daily life activities. The decision support system grounds on three-way decisions and situation awareness. An evaluation of the system following the SAGAT approach has been done and reported in the paper.

Application of Granular Computing and Three-way decisions to Analysis of Competing Hypotheses

We present an application of Granular Computing and Three-way decisions to intelligence analysis. In particular we extend the Analysis of Competing Hypotheses with an additional perspective devoted to support analysts in reasoning with groups of hypotheses that can be equivalent on the basis of partial and incomplete evidence, and in classifying these groups of hypotheses with respect to a decisional attribute of interest for the analyst, such as dangerous or safe. Creating and reasoning with granules and multi-level granular structures give to our approach an added value when dealing with a large number of evidence and hypotheses. Three-way decision making offers the possibility of a rapid understanding of how granules of hypotheses approximate a class of dangerous hypotheses, with clear benefits when analysts have to take decision on classifying a group of hypotheses or setting a proper level of attention to group of equivalent hypotheses.

Integrating GSO and SAW ontologies to enable Situation Awareness in Green Fleet Management

The definition of a Situation Awareness system is a challenging task that must be guided by some design principles like, for instance, organizing information around goals and supporting trade-off between goal-driven and data-driven information processing. This paper reports the results about the definition of a system for Green Fleet Management that synergistically exploits goal-driven and data-driven information processing to support Situation Awareness. These results are achieved by defining a framework based on the integration of several existing ontologies, which formalize and link actionable knowledge on goals and situations, with agents able to perform inference on it. Such a framework, which is generally applicable to heterogeneous domains, is also one of the main results of this work. The system has been evaluated by using the SAGAT method.

Collective awareness in Smart City with Fuzzy Cognitive Maps and Fuzzy sets

We present a methodology to support urban planners and decision makers in obtaining a good awareness of how city assets (points of interest) are perceived by a community, and on the impact and influence that this collective perception can have on other city assets and city issues such as mobility, environment, security. The methodology employees Fuzzy Cognitive Maps and Fuzzy sets. Fuzzy Cognitive Maps are used to model the relationships between elements of mental representations that different communities have with regards to city issues. The concept of signature as relation between two fuzzy sets is adopted, in analogy to what proposed by Yager and Reformat [1], to characterize a point of interest. Different signatures are subsequently grouped to characterize an area and adopted, in combination with sentiment analysis, to derive a measure of collective perception on the quality of the area. This measure is used to activate some qualitative concept of a Fuzzy Cognitive Map and perform what-if analysis. The methodology has been applied to a sample of three POIs (representing three attractions of the city of Salerno) by using data gathered from the Web and involving some real citizens. Our preliminary results are encouraging with regards to the possibilities offered by our approach of enforcing city decision makers with a good awareness on how changes in the perception of quality of urban areas can influence other city related issues.

Handling Continuity in Seamless Learning via Opportunistic Recognition and Evaluation of Activity Cohesion

Handling the continuity of learning experience across different activities and contexts is a key challenge for seamless learning. Current context and activity recognition techniques work well in fixed environments where sensors deployment and data are known but are not adaptable to dynamic and changing situations when, for instance, a learner moves from dense to rare sensor environments. Moreover, even if we are able to recognize with more or less precision activities, it still remains the issue of understanding if there are useful and interesting educational concepts related to the activities. In this short paper we discuss our ideas and preliminary results on the definition of an opportunistic approach to recognize activities and contents that leverages on the characterization of the environments in terms of sensor richness and knowledge expressiveness. The basic idea is to recognize the kind of environment in which a learner is involved and then to adapt the most suitable techniques taking advantage of the specific features of the environment. Next, we discuss two measures allowing us to understand i) the cohesion degree of the set of (informal, not formal, formal) activities a learner is involved in, and ii) if the the learner is able and in a proper disposition to acquire new knowledge or develop a new competence from the execution of activities. We propose the adoption of the first measure in the fitness function of a swarm intelligence algorithm to optimise the search of cohesive activities.

Enhancing an AmI-Based Framework for U-commerce by Applying Memetic Algorithms to Plan Shopping

Thanks to the phenomenal proliferation of the electronic commerce, the number of Internet shops increases more and more each year. This increasing forces strong competition on the market by leading to low prices for customers, but, at the same time, it represents a problem for customers since it makes difficult to manually compare all the product offers and decide a shopping plan. This scenario is furthermore made complex by a recent business strategy adopted in e-commerce scenario: the loyalty program such as point systems and coupons. In order to face the shopping plan problem in these new loyalty program scenarios, a recently proposed AmI-based framework for u-commerce introduces the exploitation of evolutionary algorithms, and, in particular, genetic ones. However, in spite of their successfully application to several complex problems, genetic algorithms are inherently characterized by premature convergence. Therefore, this paper proposes to replace the exploited evolutionary approach with the application of memetic algorithms for solving the shopping plan problem. As shown by a statistical test, our approach significantly improves the above AmI-based framework for u-commerce.

Employing Fuzzy Consensus for Assessing Reliability of Sensor Data in Situation Awareness Frameworks

Situation identification is a complex task that is usually employed in order to sustain the work of Decision Support Systems in several and heterogeneous application scenarios like, for instance, Emergency Management, Safety and Security. Typically, situation awareness systems gather and process raw sensor data by means of different techniques. In this context, it is fundamental to exploit qualitative sensor data in order to guarantee the reliability of the situation identification task results. The consolidation of Internet of Things and the growth of the Linked Sensor Data ecosystem provide us with different degrees of availability and, sometimes, redundancy of sensor observations that could be conflicting. This could be caused by sensor failures due to contextual factors, malicious attacks, faults. This paper proposes an approach based on Fuzzy Consensus to assess data coming from a group of redundant sensors and provide reliable observations to be exploited for situation identification. Lastly, Granular Computing paradigm is adopted to handle multigranularity of information, i.e., To manage observations assessed in different linguistic term sets.