Visual sentiment analysis is a challenging problem. Many datasets and approaches have been designed to foster breakthroughs in this trending research topic. However, most works scrutinize only subsymbolic models through visual attributes of the evaluated images, paying less attention to the subjectivity of viewers’ perceptions as a basis for neuro-symbolic systems. Aiming to fill this gap, we present PerceptSent, a novel dataset for visual sentiment analysis that spans 5,000 images shared by users on social networks. Besides the sentiment opinion (positive, slightly positive, neutral, slightly negative, negative) expressed by every evaluator about each image analyzed, the dataset contains evaluator's metadata (age, gender, socioeconomic status, education, and psychological hints) as well as perceptions observed by the evaluator about the image — such as the presence of nature, violence, lack of maintenance, etc. Deep architectures and different problem formulations are explored using our dataset to combine visual and extra attributes (external knowledge) for automatic sentiment analysis. We show evidence that evaluator's perceptionss, when correctly employed, are crucial in visual sentiment analysis, improving the F-score performance from 61% to an impressive rate above 97%. Although, at this point, we do not have automatic approaches to capture these perceptions, our results open up new investigation avenues.

Having the proper incentive mechanism is paramount for the success of location-based social networks (LBSNs). With that in mind, we performed a cross-cultural study on the mechanism of incentive Mayorship, which Foursquare-Swarm employs to engage its users. The user who has the mayorship is the one who performed the highest number of check-ins in the last thirty days in a particular venue. We study how alternations and disputes for mayorship occur at the venues through automatic temporal monitoring. We collected data in two cities in different countries: Curitiba (Brazil) and Chicago (United States). We found, for instance, that renowned American food chains may have a more significant influence on the mayorship dispute in Curitiba. Also, hidden prestige associated with more affluent areas could be an extra motivator factor in Curitiba, but the same cannot be said for Chicago. This study shed some light on the mechanism of incentive mayorship in different venues, showing that local factors could play an important role. Our results can assist in improving user engagement on social web systems in different cultures.

The use of contextual data to suggest distinct types of routes helps to understand new aspects of a city that may change the perception of drivers about routes. The impact of these aspects may differ from driver to driver requiring a way to change the suggestion according to the driver's point of view. Therefore, this paper presents an approach that identifies distinct situations in multiple types of contextual data and proposes a personalized and context-aware vehicle rerouting service called PONCHE. It considers common characteristics found in every dataset of spatiotemporal data to overcome the necessity of processing specific aspects of distinct data types. Regarding personalized service, each driver's profile is reflected into contextual data type weights considered by the system, i.e., the intensity he/she wants to avoid a contextual region. With that, a driver's profile may ignore a determined contextual data type. Performance evaluation results show that PONCHE identifies the best routes according to the weights given by a driver. It also improves the quality of contextual information obtained according to traffic, crime, and vehicle crashes. This study takes into consideration contextual data from Austin and Chicago in the USA, enabling comparison with two distinct cities.

Opinion mining in outdoor images posted by users during different activities can provide valuable information to better understand urban areas. In this regard, we propose a framework to classify the sentiment of outdoor images shared by users on social networks. We compare the performance of state-of-the-art ConvNet architectures and one specifically designed for sentiment analysis. We also evaluate how the merging of deep features and semantic information derived from the scene attributes can improve classification and cross-dataset generalization performance. The evaluation explores a novel dataset—namely, OutdoorSent—and other publicly available datasets. We observe that the incorporation of knowledge about semantic attributes improves the accuracy of all ConvNet architectures studied. Besides, we found that exploring only images related to the context of the study—outdoor, in our case—is recommended, i.e., indoor images were not significantly helpful. Furthermore, we demonstrated the applicability of our results in the United States city of Chicago, Illinois, showing that they can help to improve the knowledge of subjective characteristics of different areas of the city. For instance, particular areas of the city tend to concentrate more images of a specific class of sentiment, which are also correlated with median income, opening up opportunities in different fields.

Tourism has become a global economic force; for this reason, offer better services for tourists is indispensable. We explore social media data to study how tourists move through time and space and the factors that influence their movements in four major cities worldwide. In addition, we propose a new approach, based on a topic model, that enables the automatic identification of mobility pattern themes, which, ultimately, leads to the better understanding of the profile of users. Our results have implications in several segments, enabling better services for tourists in different locations that can be culturally distinct.

Some location-based social networks (LBSNs) provide, besides other spatiotemporal data, the category of venues where the data was shared from. This information allows a wide range of semantic analyses which are very useful to understand city dynamics and urban social behavior. Despite being strategic to the study of cities and societies, some LBSNs do not offer the category of venues by default. In this study, we propose an approach to identify the category of venues of unlabeled check-ins according to their geographic locations. This new classification approach is inspired by the classic k-nearest neighbor algorithm improved by mobility pattern information captured through the user's transition information observed in LBSN data. The performance evaluation of the proposed approach is performed with real-world data from different cities: London, New York City, and Tokyo. Experiments show that, for all cities, we can achieve better performances when users' mobility is taken into account. Besides, we have an indication that transfer learning regarding mobility patterns could be feasible between similar cities.

In vehicular ad hoc networks (VANETs), vehicle-to-vehicle (V2V) communication occurs opportunistically due to frequent node mobility and intermittent contact time. In this scenario, the performance evaluation of forwarding protocols by the use of the existing resources in the network is an open challenge, given the different strategies that the routing protocols adopt for choosing the next hop. Through the data analysis of three real taxis’ mobility traces and varying the radio signal range from 50 meters to 1.000 meters, we could contribute to the analysis of the existence of single-radio transmission opportunities and their quantification and classification considering either serial or parallel. Additionally, the inventory analysis of communication resources is done by evaluating the data transmission rate and the beacon overhead impact and by the proposal of a new metric (TOppMi) for the performance evaluation of forwarding protocols. We discuss the impact on forwarding protocols by the appropriate use of this metric, and we show that this metric can be used either for performance evaluation of forwarding protocols or to improve the quality over the consumption of resources in vehicular ad hoc networks. Our metric can obtain the maximum theoretical resource usage independent of the scenario. Either for Rome or San Francisco or Shanghai, as the radio’s signal range increases, the maximum theoretical amount of resources also increases. We could also show that, for the three scenarios, the beacon overhead has an insignificant impact over the total theoretical data inventory available. Furthermore, we classify the vehicles according to the contact time between them.

An increasing number of users have been adopting route recommendation systems, mostly motivated by the convenience that those systems bring to their traffic experiences. Usually, those systems observe the historical and current traffic conditions in order to evaluate and recommend the fastest routes. However, besides mobility aspects, more contextual information such as unplanned street events and neighborhood safety, are not taken into account in the recommendation process. With this in mind, we propose a platform to support context-aware route recommendation systems. The proposed platform aims to improve existing recommendation algorithms or enable the proposal of new ones. To assess it, we use datasets of routes suggested by Google Maps in the city of Curitiba, Brazil, official open data provided by the city and also data generated voluntarily by citizens in a participatory sensing fashion. Our results show the existence of an opportunity for route planners to provide personalized services to users, which is an important step towards the development of context- aware vehicular networks. Besides, these results illustrate how publicly available big data can be explored to improve context-aware route recommendations.

Learning about people's perception that emerges from urban areas has been an interesting multidisciplinary research goal because it has a great potential to ease the hard task of understanding intrinsic characteristics of urban areas. To this end, we propose an approach that explores spatial and semantic aspects in free-text messages shared on location-based social networks (LBSNs) for uncovering and mapping the perception reflected regarding urban outdoor areas. Studying outdoor areas of Chicago, we show that LBSN data carry valuable information about places and could also be used to extract urban perception, helping to better understand urban areas from many aspects. We demonstrate, through a survey with volunteers, that our approach has the potential to correctly capture the opinion considered by the users regarding the reflected perception of those areas, indicating that it could be a feasible alternative for the task under study.

We live today in a highly competitive world in many areas, for instance, in the business world. Thus, strategic business partnerships become important for a business to succeed, especially for those exposed to a dynamic and uncertain market. Therefore, this study aims to identify expressive virtual relationships among businesses that are generated spontaneously by users in social media. To this end, a business similarity model based on common spontaneous reactions of users in social media, particularly Facebook, is proposed, as well as a new algorithm for detecting business communities within the proposed model. To demonstrate the proposed approach, the study uses more than 280 million user reactions on Facebook about businesses in Curitiba, Brazil. Exploring these data and the proposed approach, this study shows the possibility to identify non-trivial relations between businesses, which might enable the development of new mechanisms for recommending strategic business partnerships for entrepreneurs and business owners.

People-centric sensing is a research topic that aims to obtain and analyze urban data from crowdsourcing, such as participatory and opportunistic sensing. Data provided by these sources increase our knowledge about different aspects of our lives in urban scenarios, which can help us to understand and address issues that cities face. Thus, the sustainable people participation is crucial to the development of this sensing paradigm. In this direction, we focus on a central element for the deployment of people-centric sensing applications: guarantee sustainable participation of users. For this, we discuss the existing challenges at the main components of an architecture to support people-centric sensing. In order to enrich this discussion, we also evaluate the incentive mechanisms used by Foursquare, mechanisms that could be used, with proper adaptation, in several types of sensing systems. Among the results, we found evidence that a specific type of incentive (mayorship-based) could be very effective to increase users' engagement. Moreover, we present a set of policies to be incorporated into an existing or new people-centric sensing architecture to complement traditional incentive mechanisms.

Smart cities emerge in computer science as a topic to cover how the technology of information and communication can be used in the urban centers to monitor its dynamics and allow the improvement of services for the citizens. In these urban centers, different methodologies are used in order to collect data and provide them to applications. These data come from several heterogeneous sources, thus there is an effort to integrate and standardize them before their use. Also, a significant amount of this data has spatio-temporal annotations, which may be used to analyze the city dynamics, such as the mobility flow. Due to these characteristics of the data generated in urban centers, and also the possibilities brought by their use and analyses, this work presents a novel approach to collect, integrate and perform some analysis tasks in mobility data from smart cities. Thus, the SMAFramework can analyze mobility patterns based on a Multi-Aspect Graph (MAG) data structure. To show the potential of the framework, it is proposed a method to analyze the saptio-temporal correlation between data from two different data sources in the same city. Real data collected from social media and a taxi system of the city of New York are used to evaluate this method. The obtained results allowed to understand some of the applicabilities of the framework and also provided some insights on how to use the framework to resolve specific problems when analyzing mobility in urban environments.

Predicting the traffic jam in urban areas is a challenge, specially when the goal is to perform short-term forecasting. We can find in the literature some advances in algorithms and techniques to handle this issue, but there is still room for innovative solutions. For example, new approaches considering different sources of information about city dynamics and urban social behavior. In fact, one of the goals of this paper is to show the benefits of using this type of data to improve short-term traffic prediction. This paper propose STRIP, a novel short-term traffic prediction model that combines logistic regressions with two urban data sources: historical data of traffic flow obtained from online maps, such as Bing Maps, and users' check-ins, shared on participatory sensor networks, which capture the routines of city inhabitants (here known as social sensors). Simulation results show that STRIP improves the accuracy of state of the art studies, specially when using data from social sensors as input.

We here investigate the potential of participatory sensor networks derived from location sharing systems, such as Foursquare, to understand the human dynamics of cities. We propose the City Image visualization technique, which builds a transition graph mapping people's movements between location categories, and demonstrate its use to identify similarities and differences of human dynamics across cities by clustering cities according to their citizens' routines. We also analyze centrality metrics of the transition graphs built for different cities, considering transitions between specific venues. We show that these metrics complement the City Image technique, contributing to a deeper understanding of city dynamics.

Participatory sensing systems (PSSs) have the potential to become fundamental tools to support the study, in large scale, of urban social behavior and city dynamics. To that end, this work characterizes the photo sharing system Instagram, considered one of the currently most popular PSS on the Internet. Based on a dataset of approximately 2.3 million shared photos, we characterize user's behavior in the system showing that there are several advantages and opportunities for large scale sensing, such as a global coverage at low cost, but also challenges, such as a very unequal photo sharing frequency, both spatially and temporally. We also observe that the temporal photo sharing pattern is a good indicator about cultural behaviors, and also says a lot about certain classes of places. Moreover, we present an application to identify regions of interest in a city based on data obtained from Instagram, which illustrates the promising potential of PSSs for the study of city dynamics.

Cities are not identical and evolve over time, and sensing in large scale can be used to capture these differences. Research in Wireless Sensor Networks has provided several tools, techniques and algorithms to solve the problem of sensing in restricted scenarios (e.g., factory). However, sensing large scale areas, such as big cities, brings many challenges and incurs high costs related to system building and management. Thus, sensing those areas becomes more feasible when people collaborate among themselves using their portable devices, and building what has been named participatory sensing systems. This work analyzes an emerging type of network derived from this type of system, the Participatory Sensor Network (PSN), where nodes are autonomous mobile entities and the sensing depends on whether they want to participate in the sensing process. Based on four datasets of participatory sensing systems (27 million of records), we show that this type of network brings many challenges related to structural problems, e.g. instant coverage very limited, and also because of big data issues, which may restrict the use of this emerging type of network. However, it presents also, as shown here, many advantages and open opportunities, mainly related to large scale study of cities dynamics.

Services that offer users the possibility of transmitting their own live streaming video content, using Web 2.0-based platforms, are increasing in popularity. In this context, we propose and evaluate solutions that contribute to improve the scalability of this type of system. This work encompasses two major steps. First, we collected data from a popular online live video sharing service, and provide a characterization of key aspects of user dynamic behavioral patterns. Next, we used our characterization findings to drive the design and evaluation, via simulation, of alternative content distribution architectures. In particular, motivated by some of our findings, we combine the traditional Peer-to-Peer and client-server architectures into a new hybrid strategy, evaluating its cost-effectiveness in comparison with the two more conventional schemes. Simulation results, covering different metrics and scenarios, indicate that the hybrid strategy yields the best tradeoffs between quality of service and bandwidth requirements.