Francesco Barile

Combining Users and Items Rankings for Group Decision Support

Traveling and city sightseeing are, in most cases, activities that involve small groups of users. Hence, a content personalization process, in a travel domain, requires taking into account simultaneously the preferences of different users. Moreover, a group recommendation system should also capture the possible intra-group relationships, which are fundamental features in a group decision process. In this paper, we model this problem as a multi-agent aggregation of preferences by using weighted social choice functions. In this context, weights can be extracted by analyzing the interactions of the group’s members on Online Social Networks.

Pre-trip Ratings and Social Networks User Behaviors for Recommendations in Touristic Web Portals

Decision-making activities in planning a city visit typically include a pre–visit hunt for information. Hence, users spend the most of the time consulting web portals in the pre–trip phase. The possibility of obtaining social media data and providing user-generated content are powerful tools for help users in the decision process. In this work, we present our framework for profiling both single users and group of users that relies on a not intrusive analysis of the users’ behaviors on social networks/media. Moreover, the analysis of the behavior of small close groups on social networks may help an automatic system in the merge of the different preferences the users may have, simulating somehow a decision process similar to a natural interaction. Such data can be used to provide POI filtering techniques on city touristic portals.

A comparison of two preference elicitation approaches for museum recommendations: A comparison of 2 preference elicitation approaches for museum recommendations

Recommendation systems based on collaborative filtering methods can be exploited in the context of providing personalized artworks tours within a museum. However, to be effectively used, we have several problems to be addressed: user preferences are not expressed as rating and recommendation systems must provide for new users efficient and simple preferences elicitation processes that do not require much effort and time. In this work, we present and evaluate 2 state‐of‐the‐art approaches that share the aim not to rely on individual item ratings. The first method uses a clustering algorithm to categorize items and provide recommendations, while the second one is inspired by the matrix factorization approach to select a couples of item groups that users have to evaluate to obtain preference profiles. We evaluate the 2 approaches with both an off‐line simulation and a user study with the aim to find the optimal configuration as well as to evaluate the effectiveness of the 2 proposed methods. Results show that the elicitation processes permit to obtain preference profiles in a time substantially less than the baseline method, while the differences in terms of prediction accuracy are minimal.

ICT Solutions for the OR.C.HE.S.T.R.A. Project: From Personalized Selection to Enhanced Fruition of Cultural Heritage Data

This work discusses the ICT solutions designed and developed within the OR.C.HE.S.T.R.A. Project. The mission of such an industrial and experimental project (Organization of Cultural Heritage and Smart Tourism and Real-time Accessibility) consists in developing some technological solutions for tourists and inhabitants aimed at appraising the cultural heritage of the historic centre of Naples. The project is based on a Social Innovation approach where services are created engaging all the possible actors in an ecosystem oriented to Smart Culture and tourism (companies, research groups and final users). Thus, in this work some innovative solutions in the cultural heritage domain are promoted and described in order to improve at the same time both the cultural knowledge to offer to different kinds of users (for instance tourists, citizens and researchers) and its learning and its preservation and protection as well. More in detail, we describe how our developed system is able to assist users before visiting the city, by suggesting them the most interesting places to see according to their preferences, and during the visit as well, in order to make the trip more interactive and enjoyable.

Evaluating the Social Benefit of a Negotiation–Based Parking Allocation

Smart parking systems usually support drivers to select parking spaces according to their preferences among competitive alternatives, which are well known in advance to the decision maker, but without considering also the needs of a city. In this paper a decision support system for selecting and reserving optimal parking spaces to drivers is presented, where the concept of optimality is related to the city social welfare including the level of satisfaction of both drivers and the city. It relies on an automated software agent negotiation to accommodate the different needs coming from the different actors involved in the parking allocation process. A simulator of such a system is evaluated with respect to a case of complete information sharing among agents, and a case of no shared information. Different metrics to evaluate the social benefit of the parking allocation in terms of both agents utilities, and allocation efficiency are considered.

Recommendation in museums: paths, sequences, and group satisfaction maximization

This work addresses the problem of generating and then recommending an artworks sequence for a group of visitors within a museum. Differently from a recommender system for an e-commerce application, the problem, here, is trying to maximize the satisfaction of the proposed recommendations, while taking into account an items’ ordering that satisfies each group member during the sequence and the artworks locations in the museum. Moreover, since many visitors may not be able to visit every artwork, the recommender system should provide suggestions while satisfying temporal visit constraints. The problem formulation is discussed together with the characteristics of a feasible solution. An exact search algorithm from the literature is used to efficiently solve the problem and to define the prerequisites for the recommender system. Finally, we evaluate a prototype implementation with both an offline analysis and a pilot study in a simulated museum environment.

Negotiating and Executing Composite Tasks for QoS-Aware Teams of Robots

The problem of allocating tasks to a team of robots composing a complex activity with global performance constraints to be met, is NP-hard. Automated negotiation was proposed as a viable heuristic approach allowing for the dynamic adjustment of the performance levels provided by the single robots in the case of robots with limited resources. This approach leads to an improved exploitation of robots capabilities in terms of the number of composite activities that can be successfully allocated to the team. In the present work, the proposed approach is extended to include the possibility for the robots to negotiate for task allocation, and to execute the tasks in an interleaved way, so that the capabilities of the entire team can be better exploited, reducing the time the robots are inactive.

A Multi-agent System for Group Decision Support Based on Conflict Resolution Styles.

With the pervasive use of social networks supporting digital communities, the problem of finding a solution to a given problem shared by a group of users that meets the requirements/preferences of its members is gaining great interest in several research domains. Software systems supporting the decision-making process taking place when building “group solution” would greatly enhance the potentiality of these digital communities. In the present work, a Group Decision Support System is proposed to help a group of users to find a set of tourist attractions, selected among a huge set of possible alternatives, that meets the preferences of each individual. The proposed system relies on an automatic negotiation mechanism to incrementally build a single recommendation for the whole group, according to the individual lists of preferred attractions of each member. Negotiation occurs among software agents that simulate different conflict resolution styles of the real users they respectively represent. Experimental results show the effectiveness of the system also when dealing with real end users preferences.

User and Group Profiling in Touristic Web Portals Through Social Networks Analysis

Touristic Web Portals can be considered windows on cultural cities. By providing all the necessary information in one single portal, the user is free to decide her/his preferred items/activities without the need of consulting different information sources. However, this kind of interface introduces the typical information overload problem. In this work, we present our framework for profiling both a single user and a group of users that relies on a not intrusive analysis of the users’ behaviors on social networks/media. By using data drawn from social networks, it is possible to obtain useful indirect information to profile occasional users. Moreover, the analysis of the behavior of small close groups on social networks may help an automatic system in the merge of the different preferences the users may have, simulating somehow a decision process similar to a natural interaction. In this direction, our aim is to identify key users taking in account concepts from research on users’ connectivity and on users’ communication activity.

An Altruistic-Based Utility Function for Group Recommendation

Preference aggregation strategies, that are inspired by economic models of decision makers, typically assume that the individual preferences of the group members depend only on their own individual evaluations of the considered items. In this direction, group recommendation algorithms rely on such standard aggregation techniques that do not consider the possibility of evaluating social interactions and influences among group’s members, as well as their personalities, which are, indeed, crucial factors in the group’s decision-making process, especially regarding small groups. On the contrary, the laboratory data have encouraged the development of models of other-regarding preferences since altruism, fairness, and reciprocity strongly motivate many people. In this paper, starting from a utility function from the literature, which combines the user personal evaluation of an item with the ones of the other group members, we propose a group recommendation method that takes into account altruism. Such function models the level of a user’s altruistic behavior starting from his/her agreeableness personality trait. Once such utility values are evaluated, the goal is to recommend items that maximize the social welfare. Performance is evaluated with a pilot study and compared with respect to Least Misery. Results showed that while for groups of two people Least Misery performs slightly better, in the other cases the two methods are comparable.