Abstract
Agent-based modeling (ABM) is a bottom-up modeling approach, where each entity of the system being modeled is uniquely represented as an independent decision-making agent. Large scale emergent behavior in ABMs is population sensitive. As such, the number of agents in a simulation should be able to reflect the reality of the system being modeled, which can be in the order of millions or billions of individuals in certain domains. A natural solution to reach acceptable scalability in commodity multi-core processors consists of decomposing models such that each component can be independently processed by a different thread in a concurrent manner. In this paper we present a multithreaded Java implementation of the PPHPC ABM, with two goals in mind: (1) compare the performance of this implementation with an existing NetLogo implementation; and, (2) study how different parallelization strategies impact simulation performance on a shared memory architecture. Results show that: (1) model parallelization can yield considerable performance gains; (2) distinct parallelization strategies offer specific trade-offs in terms of performance and simulation reproducibility; and, (3) PPHPC is a valid reference model for comparing distinct implementations or parallelization strategies, from both performance and statistical accuracy perspectives.

Abstract

Abstract
This paper describes a proposal to develop a Tourism Recommendation System based in Users and Points-of-Interest (POI) functionality/accessibility levels. The focus is to evaluate if user’s physical and psychological functionality levels can perform an important role in recommendation results accuracy. This work also aims to show the importance of POI classification (accessibility levels are related with each POI ability to receive tourists with certain levels of physical and psychological issues), through the definition of a different model regarding their accessibility and other characteristics. © Springer International Publishing AG 2018.

Abstract
The application of statistics has proliferated in recent years and has become increasingly relevant across numerous fields of study. With the advent of new technologies, its availability has opened into a wider range of users. Comparative Approaches to Using R and Python for Statistical Data Analysis is a comprehensive source of emerging research and perspectives on the latest computer software and available languages for the visualization of statistical data. By providing insights on relevant topics, such as inference, factor analysis, and linear regression, this publication is ideally designed for professionals, researchers, academics, graduate students, and practitioners interested in the optimization of statistical data analysis.

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