Maria Laterza

A recommendation method for e-learning environments: the rule-based technique

The ever growing importance of e-learning over the last decade has triggered an explosion of resources available on the Web. Although this multiplicity of available resources can foster the development of a critical spirit, discernment and the student’s ability to weigh up the merits of different points of view, it can also induce disorientation in the search for the resources best suited to her/his needs and learning style. This issue has driven research into recommendation systems, already well known in fields such as e-commerce, applied to e-learning environments. However, in such environments the recommendations can only be efficacious if the system is able to deal with the many different factors involved in the learning process, such as the learning goal and the student’s cognitive features. This work proposes a recommendation strategy that combines, by adopting a hybrid cascade approach, two knowledge-based techniques that can take these factors into account in the recommendation process.

Automated generation of itineraries in recommender systems for tourism

Current recommender systems can support tourists in choosing travel products (accommodation, activities, means of transport, etc.), in planning long trips, and in profitably spending time in a specific geographical area such as a region (or a city). In the last case, the system should be able to construct itineraries suited to the tourists interests. In this paper, a method for generating tourist itineraries in knowledge-based recommender systems is proposed. The method is based on a theoretical model that defines space-time relations among items of intangible cultural heritage (called events) and on transitive closure computation (of the relations), that is able to construct chains of events. The proposed method has been implemented in the T-Path recommender system, that suggests itineraries of cultural events occurring in the Apulia region.

A model for generating tourist itineraries

A challenge in the recommender systems currently available for the tourism domain is how to suggest tourist itineraries in a specific geographical area (city or region). The proposed theoretical model allows items of intangible cultural heritage (events) such as processions, festivals, special markets, etc. to be characterized and correlated. The model features both a set of functions characterizing the events and a space-time relation that defines whether two events are correlated. The model allows itineraries to be constructed by computing the transitive closure of the space-time relation on the set of events. It can be used to construct itineraries at different grain sizes. This capacity makes the model scalable and easily applicable in the development of several applications. It has been implemented in a first order logic knowledge base in order to make an empirical evaluation of the model.

A recommendation system to promote local cultural heritage

The value of cultural heritage is widely recognised both at national and international level, because it allows the cultural roots, memories and identity of a territory to be discovered. For this reason it is not limited to historic sites, monuments or churches, but must necessarily include a knowledge of the traditions, legends and religious rites which are part of the culture of a place. Web portals for the promotion of tourism usually offer information limited to only one category of items and this information is not sufficient to gain a deeper knowledge of a place or event. Moreover, it is rare to find on the Net solutions that are able to support teachers choosing the sites to be visited in an educational tour. The paper describes a recommender system for cultural heritage that is able to support both tourists and teachers selecting items (tangible or intangible) to visit, and that is able to offer in-depth material, selected according to the interests of the target users, that can consolidate the knowledge of the places visited. The main features of the system are: the use of a user-centred and collaborative approach to promote knowledge; a set of metadata that allows the resources to be contextualised in the culture of a territory; the use of a recommender method that takes into account the user’s preferences, multi-criteria user feedbacks and the semantic relationships among the items.

Recommendation of Collaborative Activities in E-learning Environments

In distance education environments, collaborative activities such as wikis, forums and chats play an important role in the e-learning experience because they promote communication among students and so allow cooperative learning settings to be implemented. Nevertheless, it could be difficult for learners to pick out the most interesting and appropriate collaborative activities to meet their learning needs. Recommender systems integrated in e-learning platforms are usually used mainly to help learners choose teaching resources, but they can also be useful to suggest the collaborative activities that best fit their learning objectives from a pedagogical point of view. In this context, the paper presents a recommendation approach able to suggest collaborative activities such as forums, chats, wikis and blogs, that combines dynamic clustering and prediction calculus on the basis of the learners’ profiles and needs.

A semantic approach implemented in a system recommending resources for cultural heritage tourism

In the last few years the tourism industry has profoundly changed. Today, a large number of tourists use the internet to find destinations, itineraries, services or travel packages, rather than asking experts in the field. For this reason, Information and Communication Technologies play an important role in tourism promotion. In particular, recommendation systems are interesting because they are able to offer more appropriate support than traditional search engines to the user seeking places to visit. Recommendation systems, in fact, are able to suggest a personalised set of options according to the user’s needs and preferences. But, as widely recognized in the literature, with these systems the quality of the recommendation is closely linked to the description of both resources and users. For them to become effective tools for promoting the knowledge, culture and traditions of a territory, it is necessary to integrate semantic information into the descriptions of resources, that can capture relationships among them. In this way, it is possible to enrich the list of recommendations by adding those resources which, although not explicitly related to the users request, have some semantic relationships with those included in the list. This can help to promote the discovery of new scenarios and the spread of knowledge about the cultural heritage of a territory. In this scenario, the paper presents a semantic approach amplifying cultural resources recommendations. This approach was used to enrich the list of recommendations of CulTuRek, a system that promotes the exploration of tangible and intangible cultural heritage in the Apulia Region.

An evaluation method for multi-agent systems

The growing employment of Multi-Agent Systems (MASs) in several domains of everyday life has provided the impetus for much research into new tools and methodologies for their design and implementation. But up to now, few works have focused on evaluation of these MASs, and none of these considered characteristics such as the rationality, the autonomy, the reactivity and the environment adaptability of the agents in the MAS. We believe these characteristics affect the whole performance of these systems and are connected to the complexity of the environment where the agents act. In this paper we propose an evaluation method for static multi-agent systems. The method, based on the Goal-Question-Metric approach, allows evaluation of these same MAS characteristics and combines two analysis perspectives of these systems: intra-agent and inter-agent. We also report the use of the defined approach to evaluate the GeCo_Automotive systems MAS.