Abstract
The proliferation of new technologies is revolutionizing the public transport sector, allowing Operators to replace complex and expensive infrastructures by travellers' mobile devices and online management channels and platforms. This paper aims to present the Seamless Mobility platform, a disruptive solution based on these new channels, which main goal is to facilitate and promote public transport usage for travellers, as well as reducing operational costs for public transport companies. The Seamless Mobility platform integrates three main components: (i) mobile payments, (ii) route planner, and (iii) social network. The payment component is based on the pay-as-you-go concept with check-in and check-out requiring the reading of the corresponding QR Code station. The route planner combines information from published schedules with real-time information to identify the nearest stops, the next departures, or the best route for a scheduled trip. The social network component allows real time sharing among travellers of public transport information, related to several aspects of the service (e.g. noise, skilfulness of drivers). To test the concept, a mobile application, called OneRide, was developed. This application was tested by users in real environment, in the city of Porto, Portugal. The results show that users considered the system extremely useful, since it is more convenient than traditional systems. It was also clear that users valued the integration of additional and complementary services with mobile payments, such as information about their journey, maps and schedules. Regarding the social component some users found it difficult to understand the concept, but once they understood they considered it very useful. The use of the QR Codes to perform the payment has shown to be one of the main challenges to be addressed, since lighting conditions, position and distance to the QR Code influences the reading process. (C) 2017 The Authors. Published by Elsevier B.V.

Abstract
Nowadays, mobile phones are ubiquitous systems of our society. Nevertheless, the adoption of this technology to perform mobile payments, namely in public transport, was only implemented in a few number of transport networks. Thus, this paper aims to understand which are the main factors that may influence the adoption of mobile payments in public transport. For this purpose, a survey was applied to different groups of population. The study was conducted on the public transport networks of a medium-sized metropolitan area (Oporto-Portugal) and of a big-sized metropolitan area (Beijing-China). The evaluation results of the current services of purchase and validation of public transport tickets almost never show significant statistical differences (p>0.05) for the traditional variables used by the literature. This is particularly true for age. Nevertheless, some mobility factors can sometimes play an important role in the assessment of ticketing systems. Moreover, although the high differences between the ticketing systems in both cities, Chinese and Portuguese have a similar opinion about the systems implemented in their cities. Still, Chinese reveal a higher motivation to adopt the new ticketing system. In general, such system is greatly accepted by the respondents and the potential market is expected to be high (30-68%). Although this technology cannot replace the traditional systems, it can contribute to increasing the overall efficiency of the transport networks. The improvement of the passengers' appraisal, the reduction of operational and the maintenance costs of transport operators are the network impacts most expected. Convenience and time saving are the main advantages mentioned while questions about privacy, interaction and reliability are stated as the main concerns to adopt it. (C) 2017 The Authors. Published by Elsevier B.V.

Abstract
This paper proposes an ontology-based approach to support the process of visualizing urban mobility data. The approach consists of building a visualization-oriented urban mobility ontology, focused on themes such as ridership, vehicle flows and the like. Existing ontologies focus on modelling the overall structure of transportation networks, and do not address the formalization of such themes. The ontology also allows characterizing visualization techniques with human perception factors, so that they can be used to automatically infer recommended techniques for a dataset. The ultimate goal is to benefit decision makers, by providing an ontology that can assist with the process of developing semantically-rich visualizations, with increased data interoperability and knowledge extraction capabilities. We provide an example with real data of the public transportation system of the city of Porto, Portugal. The example shows the semantic characterization of a visualization technique, and how semantics can assist the task of automatically recommending visualizations. (C) 2017 The Authors. Published by Elsevier B.V.

Abstract
The public transport system integrates a complex ecosystem, composed not only by transport operators and travellers but also by other services such as schools, firms, restaurants, museums, banks, and public establishments. Therefore, by adopting a holistic point of view, we propose a new service approach linking city services and public transport. This approach consists in partnerships that may include discounts, combined packages, reduced prices, deals and marketing campaigns, targeted to each specific audience. In order to develop these partnerships it is important to analyse the services located around the stations and the public transport usage. We use the city of Porto, Portugal, as an illustrative example and we rely on two data sources: Automated Fare Collection system data and business data points. The analysis of both datasets allowed us to determine the level of concentration of city services located around public transport stations and to identify the types of services that tend to agglomerate near the stations. We were also able to analyse the correlation between the number of travel card validations and the number of services located around the stations. Finally we present a case of a service exposure to different demographic segments.

Abstract
The problem of assigning students to teams can be described as maximising their profiles diversity within teams while minimising the differences among teams. This problem is commonly known as the maximally diverse grouping problem and it is usually formulated as maximising the sum of the pairwise distances among students within teams. We propose an alternative algorithm in which the within group heterogeneity is measured by the attributes' variance instead of by the sum of distances between group members. The proposed algorithm is evaluated by means of two real data sets and the results suggest that it induces better solutions according to two independent evaluation criteria, the Davies–Bouldin index and the number of dominated teams. In conclusion, the results show that it is more adequate to use the attributes' variance to measure the heterogeneity of profiles within the teams and the homogeneity among teams. © 2017 SEFI.

Abstract
Customers' response is an important topic in direct marketing. This study proposes a data mining response model supported by random forests to support the definition of target customers for banking campaigns. Class imbalance is a typical problem in telemarketing that can affect the performance of the data mining techniques. This study also contributes to the literature by exploring the use of class imbalance methods in the banking context. The performance of an undersampling method (the EasyEnsemble algorithm) is compared with that of an oversampling method (the Synthetic Minority Oversampling Technique) in order to determine the most appropriate specification. The importance of the attribute features included in the response model is also explored. In particular, discriminative performance was enhanced by the inclusion of demographic information, contact details and socio-economic features. Random forests, supported by an undersampling algorithm, presented very high prediction performance, outperforming the other techniques explored.