Smart cities are one of the emerging domains for computational applications. Many of these applications may benefit from the ubiquitous computing paradigm to provide better services. An important aspect of these applications is how to obtain data about their users and understand them. Context-aware approaches has been proven to be successful in understanding these data. These solutions obtain data from one or more sensors and apply context recognition techniques to infer higher level information. Several works in the last decade have presented ubiquitous approaches for context recognition that can be applied in smart cities. Our work presents a systematic mapping that provides an overview of context recognition approaches applied in smart cities domains. Several aspects of these approaches have been analyzed, such as reasoning techniques, sensors usage, context level, and applications. Of the total 3627 papers returned in the search, 93 papers were analyzed after two filtering processes. The analysis of these papers have shown that only few recent works explored situation recognition information and the full potential of the sensing capabilities in smart cities.The main objective of this article is the identification of future open context recognition approaches allowing the development of news solutions and research.

Abstract: Recommender systems appeared in the early 90s to help users deal with cognitive overload brought by the internet. From there to now, such systems have assumed many other roles like help users to explore, improve decision making, or even entertain. The system needs to look to user characteristics to accomplish such new goals. These characteristics help understand what the user task is and how to adapt the recommendation to support such task. Related research has proposed recommender systems in education. These recommender systems help learners to find the educational resources most fit for their needs. In this paper, we present an integration model between recommender and adaptive hypermedia systems. It results in a new process for educational resource recommendation, using a new algorithm of adaptive recommendation. Through a prototype and an online experiment on the educational scenario, we proved that AwARE could improve the recommendation accuracy, interaction with the system, and user satisfaction. Besides the prototype description, the paper presents a protocol to evaluate the proposed approach by both the providers’ and consumers’ point of view.

Abstract: Recommender systems have been constantly refined to improve the accuracy of rating prediction and ranking generation. However, when a recommender system is too accurate in predicting the users’ interests, negative impacts can arise. One of the most critical is the filter bubbles creation, a situation where a user receives less content diversity. In the news domain, such effect is critical once they are ways of opinion formation. In this paper, we aim to assess the role that a specific set of recommender algorithms has in the creation of filter bubbles and if diversification approaches can decrease such effect. We also verify the effects of such an environment in the users’ exposition and interaction to fake news in the Brazilian presidential election of 2018. To perform such a study, we developed a prototype that recommends news stories and presents these recommendations in a feed. To measure the filter bubble, we introduce a new metric based on the homogenization of a recommended items’ set. Our results show KNN item-based recommendation with the MMR diversification algorithm performs slightly better in putting the user in contact with less homogeneous content while presenting a lower index of likes in fake news.

Abstract: The emergence of automated environments to add intelligence to the decision making process on real world problems with multiple and conflicting objectives is an actual issue. In some cases, these objectives have the same importance and no easy prioritization can be done. For dealing with reasoning on every-day activities, systems for ambient intelligence (AmI) often deal with multi-objective problems. This paper presents a situation-aware model to support multi-objective decision making for objectives with equal importance in AmI. By using contextual data from the environment, a system based on this model identifies the situation of interest; performs multi-objective decision without assigning weights to the objectives, and performs an action to control the environment. To verify the model, a system was developed aiming to manage the multi-objective problem of thermal comfort and energy consumption of an office. As results, this work developed an L-fuzzy library (used in the decision module) and shows that the inclusion of this intelligence in AmI systems allows the achievement of both objectives, without the need of giving priority to one or to the other aspect.

Abstract: The selection of resources within a university campus is not an easy task, due the multitude of physical and virtual resources available. By this fact, a person can easily select a set of resources that cannot attend their needs and interests in the best way. Such multitude of resources can leads to the problem of the person cognitive overload, which happens during the resources selection. This paper presents a novel recommender approach as an alternative to select such best set of resources and avoid the person cognitive overload during it. Due to specificities for recommending learning resources, a knowledge-based approach seems to be more effective to this task and take the context in consideration can help in selecting the best resources to the current user situation. Take in consideration an adaptive interface makes the recommendations non-intrusive and help the user to see the location of the resources in the environment. To validate our approach we are developing a system to be used in an application scenario presented in this paper.