http://hdl.handle.net/10366/129327 Thu, 23 Apr 2026 00:14:09 GMT 2026-04-23T00:14:09Z http://hdl.handle.net/10366/129354 Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129354 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129353 In this paper, we assess the applicability of user simulation techniques to generate dialogs which are similar to real human-machine spoken interactions.To do so, we present the results of the comparison between three corpora acquired by means of different techniques. The first corpus was acquired with real users.A statistical user simulation technique has been applied to the same task to acquire the second corpus. In this technique, the next user answer is selected by means of a classification process that takes into account the previous dialog history, the lexical information in the clause, and the subtask of the dialog to which it contributes. Finally, a dialog simulation technique has been developed for the acquisition of the third corpus. This technique uses a random selection of the user and system turns, defining stop conditions for automatically deciding if the simulated dialog is successful or not. We use several evaluation measures proposed in previous research to compare between our three acquired corpora, and then discuss the similarities and differences with regard to these measures. Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129353 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129352 Smart cities are proposed as a medium-term option for all cities. This article aims to propose an architecture that allows cities to provide solutions to interconnect all their elements. The study case focuses in locating and optimized regulation of traffic in cities. However, thanks to the proposed structure and the applied algorithms, the architecture is scalable in size of the sensor network, in functionality or even in the use of resources. A simulation environment that is able to show the operation of the architecture in the same way that a real city would, is presented. Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129352 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129351 Most approaches to keywords discovery when analyzing microblogging messages (among them those from Twitter) are based on statistical and lexical information about the words that compose the text. The lack of context in the short messages can be problematic due to the low co-occurrence of words. In this paper, we present a new approach for keywords discovering from Spanish tweets based on the addition of context information using Wikipedia as a knowledge base. We present four different ways to use Wikipedia and two ways to rank the new keywords. We have tested these strategies using more than 60000 Spanish tweets, measuring performance and analyzing particularities of each strategy. Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129351 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129350 The growth of personal sensors and the ability to sensorize attributes connected with the physical beings and environments are increasing. Initiatives such as Internet of Things (IoT)) aim to connect devices and people through communication channels in order to automate and fuel interaction. Targeted approaches can be found on the Smart Cities projects which use the IoT to gather data from people and attributes related to city management. Though good for management of new cities, well-being should as well be of principal importance. It regards users higher than infrastructure and managerial data. Taking lessons from ergonomic studies, health studies and user habits it is possible to track and monitor user daily living. Moreover, the link between user living conditions and sparse events such as illness, indispositions can be tracked to well-being data through autonomous services. Such application is detailed in the approach categorized by this article and the research presented Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129350 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129349 Automatic keyword extraction (AKE) from textual sources took a valuable step towards harnessing the problem of efficient scanning of large document collections. Particularly in the context of urban mobility, where the most relevant events in the city are advertised on-line, it becomes difficult to know exactly what is happening in a place./nIn this paper we tackle this problem by extracting a set of keywords from different kinds of textual sources, focusing on the urban events context. We propose an ensemble of automatic keyword extraction systems KEA (Key-phrase Extraction Algorithm) and KUSCO (Knowledge Unsupervised Search for instantiating Concepts on lightweight Ontologies) and Conditional Random Fields (CRF).; /nUnlike KEA and KUSCO which are well-known tools for automatic keyword extraction, CRF needs further pre-processing. Therefore, a tool for handling AKE from the documents using CRF is developed. The architecture for the AKE ensemble system is designed and efficient integration of component applications is presented in which a consensus between such classifiers is achieved. Finally, we empirically show that our AKE ensemble system significantly succeeds on baseline sources and urban events collections. Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129349 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129348 Road traffic is an important phenomenon in modern societies. The study of its different aspects in the multiple scenarios where it happens is relevant for a huge number of problems. At the same time, its scale and complexity make it hard to study. Traffic simulations can alleviate these difficulties, simplifying the scenarios to consider and controlling their variables. However, their development also presents difficulties. The main ones come from the need to integrate the way of working of researchers and developers from multiple fields. Model-Driven Engineering (MDE) addresses these problems using Modelling Languages (MLs) and semi-automatic transformations to organise and describe the development, from requirements to code. This paper presents a domain-specific MDE framework for simulations of road traffic. It comprises an extensible ML, support tools, and development guidelines. The ML adopts an agent-based approach, which is focused on the roles of individuals in road traffic and their decision-making. A case study shows the process to model a traffic theory with the ML, and how to specialise that specification for an existing target platform and its simulations. The results are the basis for comparison with related work. Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129348 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129347 We present a project implemented on the field which has two separate strands, one refers on collecting crowd sensing data through mobile apps where context is (near) automatically induced, another is related to a practical application of this method in a real time system to manage solidarity campaigns in collecting goods. Here, we cover both parts, we applied an experimental setup and obtained results and insights in a third sector institution, Caritas Diocesana of Coimbra[1], a non-profit organization part of Caritas[2]. As main contribution, we propose a distributed architecture for Mobile Crowd Sensing able not only to allow real time inventory through simultaneous campaigns but also it gives feedback to volunteers in order to instantly acquire information about which categories of goods are more needed[1] http://www.caritas.pt/site/nacional/ Portuguese Website (last visited in October 2015)/n[2] http://www.caritas.eu/ (last visited in October 2015) Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129347 2015-05-10T00:00:00Z http://hdl.handle.net/10366/129346 Sun, 10 May 2015 00:00:00 GMT http://hdl.handle.net/10366/129346 2015-05-10T00:00:00Z