Miguel Araújo

Com2: Fast automatic discovery of temporal ('comet') communities

Given a large network, changing over time, how can we find patterns and anomalies? We propose Com2, a novel and fast, incremental tensor analysis approach, which can discover both transient and periodic/ repeating communities. The method is (a) scalable, being linear on the input size (b) general, (c) needs no user-defined parameters and (d) effective, returning results that agree with intuition.

TraSMAPI: An API oriented towards Multi-Agent Systems real-time interaction with multiple Traffic Simulators

TraSMAPI (Traffic Simulation Manager Application Programming Interface) is designed to provide real-time interaction with Traffic Simulators, collect relevant metrics and statistics, and offer an integrated framework to develop Multi-Agent Systems. It is presented as a tool for the simulation of dynamic control systems in road networks with special focus on Multi-Agent Systems. The abstraction over the simulator opens up the possibility of running different traffic simulators using the same API (application programming interface) allowing the comparison of results of the same application in different simulators. The proposed approach is, therefore, expected to be a key asset in supporting and enhancing engineers and practitioners to make more effective control decisions and implement more efficient management policies while analyzing and addressing traffic related problems in urban areas.

Beyond blocks: hyperbolic community detection

What do real communities in social networks look like? Community detection plays a key role in understanding the structure of real-life graphs with impact on recommendation systems, load balancing and routing. Previous community detection methods look for uniform blocks in adjacency matrices. However, after studying four real networks with ground-truth communities, we provide empirical evidence that communities are best represented as having an hyperbolic structure. We detail HyCoM - the Hyperbolic Community Model - as a better representation of communities and the relationships between their members, and show improvements in compression compared to standard methods. We also introduce HyCoM-FIT, a fast, parameter free algorithm to detect communities with hyperbolic structure. We show that our method is effective in finding communities with a similar structure to self-declared ones. We report findings in real social networks, including a community in a blogging platform with over 34 million edges in which more than 1000 users established over 300 000 relations.

Short-term real-time traffic prediction methods: A survey

Short-term traffic prediction provides tools for improved road management by allowing the reduction of delays, incidents and other unexpected events. Different real-time approaches provide traffic managers with varying but valuable information. This paper reviews the literature regarding model-driven and data-driven approaches focusing on short-term realtime traffic prediction. We start by analyzing real-time traffic data collection, referring network state acquisition and description methods which are used as input to predictive algorithms. According to the input variables available, we describe common and useful traffic prediction outputs that should contribute to understand the panorama verified on a road network. We then discuss metrics commonly used to assess prediction accuracy, in order to understand a standardized way to compare the different approaches. We list, detail and compare existing model-driven and data-driven approaches that provide short-term real-time traffic predictions. This research leads to an understanding of the many advantages, disadvantages and trade-offs of the approaches studied and provides useful insights for future development. Despite the predominance of model-driven solutions for the last years, data-driven approaches also present good results suitable for Traffic Management usage.

Using TraSMAPI for the assessment of multi-agent traffic management solutions

Intelligent Traffic Management is undoubtedly a promising solution to tackle modern cities’ problems related to the growth of the urban traffic volume as it is a non-invasive approach when compared to interventions to the road network structure. Among possible solutions aiming at Intelligent Traffic Management, we believe that multi-agent systems (MASs) are the most appropriate metaphor to deal with complex domains such as road networks and traffic management and control systems. However, we feel that traffic management and control, particularly intelligent traffic control, is an issue that has not yet been addressed to its full potential. Therefore, we propose using the Traffic Simulation Management API’s multi-agent framework for multi-agent simulations over multiple microscopic simulators, as a basis for the development of intelligent policies for traffic management. We present a case study in which the advantages of cross-validation using two simulators are highlighted.

On the quantification of local deformation demands and adequacy of linear analysis procedures for the seismic assessment of existing steel buildings to EC8-3

The application of performance-based design and assessment procedures requires an accurate estimation of local component deformation demands. In the case of steel moment-resisting frames, these are usually defined in terms of plastic rotations. A rigorous estimation of this response parameter is not straightforward, requiring not only the adoption of complex nonlinear structural models, but also of time-consuming numerical integration calculations. Moreover, the majority of existing codes and guidelines do not provide any guidance in terms of how these response parameters should be estimated. Part 3 of Eurocode 8 (EC8-3) requires the quantification of plastic rotations even when linear methods of analysis are used. Therefore, the aim of the research presented in this paper is to evaluate different methods of quantifying local component demands and also to answer the question of how reliable are the estimates obtained using the EC8-3 linear analysis procedures in comparison to more accurate nonlinear methods of analysis, particularly when the linear analysis applicability criterion proposed by EC8-3 is verified. An alternative methodology to assess the applicability of linear analysis is proposed which overcomes the important limitations identified in the EC8-3 criterion.

A Critical Review of European and American Provisions for the Seismic Assessment of Existing Steel Moment-Resisting Frame Buildings

ABSTRACT The publication of Part 3 of Eurocode 8 (EC8-3), dedicated to the seismic assessment of existing buildings, took place a decade ago. However, its application in engineering practice has been limited. Moreover, no studies have been conducted regarding the application of EC8-3 to steel structures. In this paper, a critical review and practical application of EC8-3 and ASCE41-13 are carried out. Issues related to the definition of the performance requirements, compliance criteria, and the consistency of the analysis procedures proposed by both standards are identified. Conceptual differences between both documents are highlighted, and several inconsistencies in EC8-3 are discussed.

Using TraSMAPI for Developing Multi-Agent Intelligent Traffic Management Solutions

Intelligent Traffic Management is undoubtedly a promising solution to tackle modern cities’ problems related to the growth of the urban traffic volume as it is a non-invasive approach when compared to interventions to the road network structure. Among possible solutions aiming at Intelligent Traffic Management, we believe that Multi-Agent Systems are the most appropriate metaphor to deal with complex domains such as road networks and traffic management and control systems. However, we feel that traffic management and control, particularly intelligent traffic control, is an issue that has not yet been addressed to its full potential. Therefore, we propose in our approach to use TraSMAPI, a tool that offers the possibility of developing real-time Multi-Agent solutions over microscopic simulators, as a basis for the development of intelligent traffic management systems aiming at the creation of revolutionary solutions in the field of traffic and transport systems.

TensorCast: Forecasting Time-Evolving Networks with Contextual Information

Can we forecast future connections in a social network? Can we predict who will start using a given hashtag in Twitter, leveraging contextual information such as who follows or retweets whom to improve our predictions? In this paper we present an abridged report of TENSORCAST, a method for forecasting time-evolving networks, that uses coupled tensors to incorporate multiple information sources. TENSORCAST is scalable (linearithmic on the number of connections), effective (more precise than competing methods) and general (applicable to any data source representable by a tensor). We also showcase our method when applied to forecast two large scale heterogeneous real world temporal networks, namely Twitter and DBLP.

TensorCast: forecasting and mining with coupled tensors

Given an heterogeneous social network, can we forecast its future? Can we predict who will start using a given hashtag on twitter? Can we leverage side information, such as who retweets or follows whom, to improve our membership forecasts? We present TensorCast, a novel method that forecasts time-evolving networks more accurately than current state-of-the-art methods by incorporating multiple data sources in coupled tensors. TensorCast is (a) scalable, being linearithmic on the number of connections; (b) effective, achieving over 20% improved precision on top-1000 forecasts of community members; (c) general, being applicable to data sources with different structure. We run our method on multiple real-world networks, including DBLP, epidemiology data, power grid data, and a Twitter temporal network with over 310 million nonzeros, where we predict the evolution of the activity of the use of political hashtags.