Search

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-10422-5_20, Proceedings of Eight International Symposium on Intelligent Distributed Computing held in Madrid in September 3-5, 2014, The application of ACO-based algorithms in data mining is growing over the last few years and several supervised and unsupervised learning algorithms have been developed using this bio-inspired approach.Most recent works concerning unsupervised learning have been focused on clustering, where ACO-based techniques have showed a great potential. At the same time, new clustering techniques that seek the continuity of data, specially focused on spectral-based approaches in opposition to classical centroid-based approaches, have attracted an increasing research interest—an area still under study by ACO clustering techniques. This work presents a hybrid spectral-based ACO clustering algorithm inspired by the ACO Clustering (ACOC) algorithm. The proposed approach combines ACOC with the spectral Laplacian to generate a new search space for the algorithm in order to obtain more promising solutions. The new algorithm, called SACOC, has been compared against well-known algorithms (K-means and Spectral Clustering) and with ACOC. The experimentsmeasure the accuracy of the algorithm for both synthetic datasets and realworld datasets extracted from the UCI Machine Learning Repository., This work has been partly supported by: Spanish Ministry of Science and Education under project TIN2010-19872 and Savier an Airbus Defense & Space project (FUAM- 076914 and FUAM-076915).

The final publication is available at Springer via http://dx.doi.org/10.1007/s10270-012-0242-3, Graph transformation is being increasingly used to express the semantics of domain-specific visual languages since its graphical nature makes rules intuitive. However, many application domains require an explicit handling of time to accurately represent the behaviour of a real system and to obtain useful simulation metrics to measure throughputs, utilization times and average delays. Inspired by the vast knowledge and experience accumulated by the discrete event simulation community, we propose a novel way of adding explicit time to graph transformation rules. In particular, we take the event scheduling discrete simulation world view and provide rules with the ability to schedule the occurrence of other rules in the future. Hence, our work combines standard, efficient techniques for discrete event simulation (based on the handling of a future event set) and the intuitive, visual nature of graph transformation. Moreover, we show how our formalism can be used to give semantics to other timed approaches and provide an implementation on top of the rewriting logic system Maude., Work partially sponsored by the Spanish Ministry, under project “Go Lite” (TIN2011-24139) as well as by the R&D programme of the Community of Madrid, project “e-Madrid” (S2009/TIC-1650). We are grateful to the anonymous reviewers, which helped in improving previous versions of the paper.

Proceedings of 9th International Conference, ANTS 2014, Brussels, Belgium, September 10-12, 2014., he final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-09952-1_11, The application of ACO-based algorithms in data mining is growing over the last few years and several supervised and unsupervised learning algorithms have been developed using this bio-inspired approach. Most recent works concerning unsupervised learning have been focused on clustering, showing great potential of ACO-based techniques. This work presents an ACO-based clustering algorithm inspired by the ACO Clustering (ACOC) algorithm. The proposed approach restructures ACOC from a centroid-based technique to a medoid-based technique, where the properties of the search space are not necessarily known. Instead, it only relies on the information about the distances amongst data. The new algorithm, called MACOC, has been compared against well-known algorithms (K-means and Partition Around Medoids) and with ACOC. The experiments measure the accuracy of the algorithm for both synthetic datasets and real-world datasets extracted from the UCI Machine Learning Repository., This work has been partly supported by: Spanish Ministry of Science and Education under project TIN2010-19872 and Savier – an Airbus Defense & Space project (FUAM-076914 and FUAM-076915).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-10840-7_35, Proceedings 15th International Conference, Salamanca, Spain, September 10-12, 2014., Mission Planning is a classical problem that has been traditionally studied in several cases from Robotics to Space missions. This kind of problems can be extremely difficult in real and dynamic scenarios. This paper provides a first analysis for mission planning to Unmanned Air Vehicles (UAVs), where sensors and other equipment of UAVs to perform a task are modelled based on Temporal Constraint Satisfaction Problems (TCSPs). In this model, a set of resources and temporal constraints are designed to represent the main characteristics (task time, fuel consumption, ...) of this kind of aircrafts. Using this simplified TCSP model, and a Branch and Bound (B&B) search algorithm, a set of feasible solutions will be found trying to minimize the fuel cost, flight time spent and the number of UAVs used in the mission. Finally, some experiments will be carried out to validate both the quality of the solutions found and the spent runtime to found them., This work is supported by the Spanish Ministry of Science and Education under Project Code TIN2010-19872 and Savier Project (Airbus Defence & Space, FUAM-076915).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-39878-0_13, Proceedings of 14th International Conference, EC-Web 2013, Prague, Czech Republic, August 27-28, 2013., Context-aware recommender systems have been proven to improve the performance of recommendations in a wide array of domains and applications. Despite individual improvements, little work has been done on comparing different approaches, in order to determine which of them outperform the others, and under what circumstances. In this paper we address this issue by conducting an empirical comparison of several pre-filtering, post-filtering and contextual modeling approaches on the movie recommendation domain. To acquire confident contextual information, we performed a user study where participants were asked to rate movies, stating the time and social companion with which they preferred to watch the rated movies. The results of our evaluation show that there is neither a clear superior contextualization approach nor an always best contextual signal, and that achieved improvements depend on the recommendation algorithm used together with each contextualization approach. Nonetheless, we conclude with a number of cues and advices about which particular combinations of contextualization approaches and recommendation algorithms could be better suited for the movie recommendation domain., This work was supported by the Spanish Government (TIN2011-28538-C02) and the Regional Government of Madrid (S2009TIC-1542)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-40643-0_5, Proceedings of 15th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2013, Madrid, Spain, September 17-20, 2013., In this paper we present a contextual modeling approach for model-based recommender systems that integrates and exploits both user preferences and contextual signals in a common vector space. Differently to previous work, we conduct a user study acquiring and analyzing a variety of realistic contextual signals associated to user preferences in several domains. Moreover, we report empirical results evaluating our approach in the movie and music domains, which show that enhancing model-based recommender systems with time, location and social companion information improves the accuracy of generated recommendations.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-39143-9_27, Proceedings of 5th International Conference, CCD 2013, Held as Part of HCI International 2013, Las Vegas, NV, USA, July 21-26, 2013, Internationalization is a common practice today in software development. In the most basic sense, internationalization is carried out by applying localization design guidelines to face language translation, icon representation, character sets and so on. However, this practice is mostly intended for design purposes, which results insufficient when applying internationalization in huge projects and, specifically, through a concrete development process. In this paper, a broader framework is provided in order to ensure internationalization through a software development process. To this end, a set of activities and sub-activities will be presented involving not only design but pre-development, analysis, implementation and evaluation issues that need to be considered for a right internationalization assurance in international software development. The idea behind is to bridge the gap between simple and usual localization activities and the user-centered software development process as internationalization assurance also helps increase the quality and usability of the software overall., This work has been supported by the founded projects TIN2011-24139, S2009/TIC-1650 and TIN2011-15009-E

Proceedings of 21th International Conference, UMAP 2013, Rome, Italy, June 10-14, 2013, The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-38844-6_42, Cross-domain recommender systems suggest items in a target domain by exploiting user preferences and/or domain knowledge available in a source domain. In this thesis we aim to develop a framework for cross-domain recommendation capable of mining heterogeneous sources of information such as semantically annotated data, user generated contents, and contextual signals. For this purpose, we investigate a number of approaches to extract, process, and integrate knowledge for linking distinct domains, and various models that exploit such knowledge for making effective recommendations across domains.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-38883-5_1, Proceedings of 6th International Conference, ICMT 2013, Budapest, Hungary, June 18-19, 2013, Several styles of model transformations are well-known and widely used, such as batch, live, incremental and lazy transformations. While they permit tackling advanced scenarios, some applications deal with models that are only available as a possibly infinite stream of elements. Hence, in streaming transformations, source model elements are continuously produced by some process, or very large models are fragmented and fed into the transformation engine. This poses a series of issues that cannot be tackled using current transformation engines. In this paper we motivate the applicability of this kind of transformations, explore the elements involved, and review several strategies to deal with them. We also propose a concrete approach, built on top of the Eclectic transformation tool., This work was funded by the Spanish Ministry of Economy and Competitivity (project “Go Lite” TIN2011-24139) and the R&D programme of the Madrid Region (project “e-Madrid” S2009/TIC-1650)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-36089-3_7, Revised Selected Papers of 5th International Conference, SLE 2012, Dresden, Germany, September 26-28, 2012, The increasing application of Model-Driven Engineering in a wide range of domains, in addition to pure code generation, raises the need to manipulate models at runtime, as part of regular programs. Moreover, certain kinds of programming tasks can be seen as model transformation tasks, and thus we could take advantage of model transformation technology in order to facilitate them. In this paper we report on our works to bridge the gap between regular programming and model transformation by enabling the manipulation of Java APIs as models. Our approach is based on the specification of a mapping between a Java API (e.g., Swing) and a meta-model describing it. A model transformation definition is written against the API meta-model and we have built a compiler that generates the corresponding Java bytecode according to the mapping. We present several application scenarios and discuss the mapping between object-oriented meta-modelling and the Java object system. Our proposal has been validated by a prototype implementation which is also contributed., Work funded by the Spanish Ministry of Economy and Competitivity (TIN2011-24139), and the R&D programme of Madrid Region (S2009/TIC-1650).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-40495-5_62, Proceedings 5th International Conference, ICCCI 2013, Craiova, Romania, September 11-13, 2013, Social Networks have become an important environment for Collective Trends extraction. The interactions amongst users provide information of their preferences and relationships. This information can be used to measure the influence of ideas, or opinions, and how they are spread within the Network. Currently, one of the most relevant and popular Social Network is Twitter. This Social Network was created to share comments and opinions. The information provided by users is specially useful in different fields and research areas such as marketing. This data is presented as short text strings containing different ideas expressed by real people. With this representation, different Data Mining and Text Mining techniques (such as classification and clustering) might be used for knowledge extraction trying to distinguish the meaning of the opinions. This work is focused on the analysis about how these techniques can interpret these opinions within the Social Network using information related to IKEA® company., The preparation of this manuscript has been supported by the Spanish Ministry of Science and Innovation under the following projects: TIN2010-19872, ECO2011-30105 (National Plan for Research, Development and Innovation) and the Multidisciplinary Project of Universidad Aut´onoma de Madrid (CEMU-2012-034)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-38679-4_9, Proceedings of 12th International Work-Conference on Artificial Neural Networks, IWANN 2013, Puerto de la Cruz, Tenerife, Spain, June 12-14, 2013, Part I, The prediction and management of wind power ramps is currently receiving large attention as it is a crucial issue for both system operators and wind farm managers. However, this is still an issue far from being solved and in this work we will address it as a classification problem working with delay vectors of the wind power time series and applying local Mahalanobis K-NN search with metrics derived from Anisotropic Diffusion methods. The resulting procedures clearly outperform a random baseline method and yield good sensitivity but more work is needed to improve on specificity and, hence, precision., With partial support from Spain's grant TIN2010-21575- C02-01 and the UAM-ADIC Chair for Machine Learning. The rst author is also supported by an FPI-UAM grant and kindly thanks the Applied Mathematics Department of Yale University for receiving her during her visits. The second author is supported by the FPU-MEC grant AP2008-00167.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-40643-0_3, Proceedings of 15th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2013, Madrid, Spain, September 17-20, 2013., Online services are usually accessed via household accounts. A household account is typically shared by various users who live in the same house. This represents a problem for providing personalized services, such as recommendation. Identifying the household members who are interacting with an online system (e.g. an on-demand video service) in a given moment, is thus an interesting challenge for the recommender systems research community. Previous work has shown that methods based on the analysis of temporal patterns of users are highly accurate in the above task when they use randomly sampled test data. However, such evaluation methodology may not properly deal with the evolution of the users’ preferences and behavior through time. In this paper we evaluate several methods’ performance using time-aware evaluation methodologies. Results from our experiments show that the discrimination power of different time features varies considerably, and moreover, the accuracy achieved by the methods can be heavily penalized when using a more realistic evaluation methodology.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-39878-0_9, Proceedings of 14th International Conference, EC-Web 2013, Prague, Czech Republic, August 27-28, 2013., In this paper we present an emotion computational model based on social tags. The model is built upon an automatically generated lexicon that describes emotions by means of synonym and antonym terms, and that is linked to multiple domain-specific emotion folksonomies extracted from entertainment social tagging systems. Using these cross-domain folksonomies, we develop a number of methods that automatically transform tag-based item profiles into emotion-oriented item profiles. To validate our model we report results from a user study that show a high precision of our methods to infer the emotions evoked by items in the movie and music domains, and results from an offline evaluation that show accuracy improvements on model-based recommender systems that incorporate the extracted item emotional information., This work was supported by the Spanish Government (TIN2011-28538-C02) and the Regional Government of Madrid (S2009TIC-1542).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-40840-3_16, Proceedings of 10th International Conference, CDVE 2013, Alcudia, Mallorca, Spain, September 22-25, 2013, Domain-Specific Modelling Languages (DSMLs) are high-level languages specially designed to perform tasks in a particular domain. When developing DSMLs, the participation of end-users is normally limited to providing domain knowledge and testing the resulting language prototypes. Language developers, which are perhaps not domain experts, are therefore in control of the language development and evolution. This may cause misinterpretations which hamper the development process and the quality of the DSML. Thus, it would be beneficial to promote a more active participation of end-users in the development process of DSMLs. While current DSML workbenches are mono-user and designed for technical experts, we present a process and tool support for the example-driven, collaborative construction of DSMLs in order to engage end-users in the creation of their own languages., This work was funded by the Spanish Ministry of Economy and Competitivity (project “Go Lite” TIN2011-24139), the R&D programme of the Madrid Region (project “e-Madrid” S2009/TIC-1650) and the European Commission under the ICT Policy Support Programme, grant no. 317859.

Proceedings of 21th International Conference, UMAP 2013, Rome, Italy, June 10-14, 2013, The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-38844-6_30, We present an emotion model based on social tags, which is built upon an automatically generated lexicon that describes emotions by means of synonym and antonym terms. Using this model we develop a number of methods that transform social tag-based item profiles into emotion-oriented item profiles. We show that the model’s representation of a number of basic emotions is in accordance with the well known psychological circumplex model of affect, and we report results from a user study that show a high precision of our methods to infer the emotions evoked by items in the movie and music domains., This work was supported by the Spanish Government (TIN2011-28538-C02) and the Regional Government of Madrid (S2009TIC-1542).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-39013-5_13, Proceedings of 9th European Conference, ECMFA 2013, Montpellier, France, July 1-5, 2013, Almira Labs is a software company that develops value-added services for the telecommunications industry. It is focused on innovative technologies that enable enterprise business and mobile and landline operators to offer next-generation voice-driven applications for all types of phones. Telephony services are built atop the proprietary Umbra framework, which is a Java API relying on the JAIN SLEE standard for event-based communication applications. This paper describes Umbra Designer, a novel graphical modelling tool for the visual development of telephony services, from which Java code for the Umbra framework is synthesized. In this way, it is easy to develop ready-to-use services, even by users not familiar with the Java API or the JAIN SLEE standard. We also report on some experiments aimed at measuring the efficiency gain derived from using the graphical tool, compared with coding directly using the Java API., This work was partially funded by the Innocash program and the project “Go Lite” TIN2011-24139 of the Spanish Ministry of Economy and Competitivity. The work was also funded by the R&D programme of the Madrid Region (project “e-Madrid” S2009/TIC-1650)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-28509-7_18, This manuscript is an extended version of the paper ‘cTag: Semantic Contextualisation of Social Tags’, presented at the 6th International Workshop on Semantic Adaptive Social Web (SASWeb 2011)., We present an algorithmic framework to accurately and efficiently identify the semantic meanings and contexts of social tags within a particular folksonomy. The framework is used for building contextualised tag-based user and item profiles. We also present its implementation in a system called cTag, with which we preliminary analyse semantic meanings and contexts of tags belonging to Delicious and MovieLens folksonomies. The analysis includes a comparison between semantic similarities obtained for pairs of tags in Delicious folksonomy, and their semantic distances in the whole Web, according to co-occurrence based metrics computed with results of a Web search engine., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02), and Universidad Autónoma de Madrid (CCG10-UAM/TIC-5877)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-34176-2_5, Revised Selected and Invited Papers, 4th International Symposium, AGTIVE 2011, Budapest, Hungary, October 4-7, 2011, Model-Driven Engineering promotes models as the principal artefacts of the development, hence model transformation techniques – like graph transformation – become key enablers for this development paradigm. In order to increase the adoption of Model-Driven Engineering in industrial practice, techniques aimed at raising the quality and productivity in model transformation development are needed. In this paper we bring elements from generic programming into graph transformation in order to define generic graph transformations that can be reused in different contexts. In particular, we propose the definition and instantiation of graph transformation templates whose requirements from generic types are specified through so-called concepts, as well as mixin layers that extend meta-models with the extra auxiliary elements needed by templates., Work funded by the Spanish Ministry of Economy and Competitivity (project TIN2011-24139) and the Region of Madrid (project S2009/TIC-1650)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-31491-9_20, Proceedings of 8th European Conference, ECMFA 2012, Kgs. Lyngby, Denmark, July 2-5, 2012, Domain-specific modelling languages are normally defined through general-purpose meta-modelling languages like the MOF. While this is satisfactory for many Model-Driven Engineering (MDE) projects, several researchers have identified the need for domain-specific meta-modelling (DSMM) languages providing customised meta-modelling primitives aimed at the definition of modelling languages in a specific domain, as well as the construction of meta-model families. In this paper, we discuss the potential of multi-level meta-modelling for the systematic engineering of DSMM architectures. For this purpose, we present: (i) several primitives and techniques to control the meta-modelling facilities offered to the users of the DSMM languages, (ii) a flexible approach to define textual concrete syntaxes for DSMM languages, (iii) extensions to model management languages enabling the practical use of DSMM in MDE, and (iv) an implementation of these ideas in the metaDepth tool., This work was funded by the Spanish Ministry of Economy and Competitivity (project “Go Lite” TIN2011-24139) and the R&D programme of the Madrid Region (project “e-Madrid” S2009/TIC-1650)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-33885-4_41, Proceedings of Information Fusion in Computer Vision for Concept Recognition at the ECCV 2012, In this paper we present a new model of semantic features that, unlike previously presented methods, does not rely on the presence of a labeled training data base, as the creation of the feature extraction function is done in an unsupervised manner. We test these features on an unsupervised classification (clustering) task, and show that they outperform primitive (low-level) features, and that have performance comparable to that of supervised semantic features, which are much more expensive to determine relying on the presence of a labeled training set to train the feature extraction function.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-33266-1_70, Proceedings of 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, In this work we will apply Diffusion Maps (DM), a recent technique for dimensionality reduction and clustering, to build local models for wind energy forecasting. We will compare ridge regression models for K–means clusters obtained over DM features, against the models obtained for clusters constructed over the original meteorological data or principal components, and also against a global model. We will see that a combination of the DM model for the low wind power region and the global model elsewhere outperforms other options., With partial support from grant TIN2010-21575-C02-01 of Spain’s Ministerio de Economía y Competitividad and the UAM–ADIC Chair for Machine Learning in Modelling and Prediction. The first author is also supported by an FPI-UAM grant and kindly thanks the Applied Mathematics Department of Yale University for receiving her during a visit. The second author is supported by the FPU-MEC grant AP2008-00167. We also thank Red Eléctrica de España, Spain’s TSO, for providing historic wind energy data

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-30476-7_3, Proceedings of 5th International Conference, ICMT 2012, Prague, Czech Republic, May 28-29, 2012, Testing model transformations poses several challenges, among them the automatic generation of appropriate input test models and the specification of oracle functions. Most approaches to the generation of input models ensure a certain level of source meta-model coverage, whereas the oracle functions are frequently defined using query or graph languages. Both tasks are usually performed independently regardless their common purpose, and sometimes there is a gap between the properties exhibited by the generated input models and those demanded to the transformations (as given by the oracles). Recently, we proposed a formal specification language for the declarative formulation of transformation properties (invariants, pre- and postconditions) from which we generated partial oracle functions that facilitate testing of the transformations. Here we extend the usage of our specification language for the automated generation of input test models by constraint solving. The testing process becomes more intentional because the generated models ensure a certain coverage of the interesting properties of the transformation. Moreover, we use the same specification to consistently derive both the input test models and the oracle functions., Work funded by the Spanish Ministry of Economy and Competitivity (TIN2011-24139) and by the R&D programme of Madrid Region (S2009/TIC-1650)

Proceedigns of 12th International Conference, EC-Web 2011, Toulouse, France, August 30 - September 1, 2011., The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-23014-1_9, We present an approach that efficiently identifies the semantic meanings and contexts of social tags within a particular folksonomy, and exploits them to build contextualised tag-based user and item profiles. We apply our approach to a dataset obtained from Delicious social bookmarking system, and evaluate it through two experiments: a user study consisting of manual judgements of tag disambiguation and contextualisation cases, and an offline study measuring the performance of several tag-powered item recommendation algorithms by using contextualised profiles. The results obtained show that our approach is able to accurately determine the actual semantic meanings and contexts of tag annotations, and allow item recommenders to achieve better precision and recall on their predictions., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02), and the Community of Madrid (CCG10- UAM/TIC-5877)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-18449-9_8, Revised Selected Papers of 7th International Workshop, AMR 2009, Madrid, Spain, September 24-25, 2009, We present here, an image description approach based on prosemantic features. The images are represented by a set of low-level features related to their structure and color distribution. Those descriptions are fed to a battery of image classifiers trained to evaluate the membership of the images with respect to a set of 14 overlapping classes. Prosemantic features are obtained by packing together the scores. To verify the effectiveness of the approach, we designed a target search experiment in which both low-level and prosemantic features are embedded into a content-based image retrieval system exploiting relevance feedback. The experiments show that the use of prosemantic features allows for a more successful and quick retrieval of the query images.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-21732-6_5, Proceedings of 4th International Conference, ICMT 2011, Zurich, Switzerland, June 27-28, 2011, Model transformation is one of the core techniques in Model Driven Engineering. Many transformation languages exist nowadays, but few offer mechanisms directed to the reuse of whole transformations or transformation fragments in different contexts. Taking inspiration from generic programming, in this paper we define model transformation templates. These templates are defined over meta-model concepts which later can be bound to specific meta-models. The binding mechanism is flexible as it permits mapping concepts and meta-models with certain kinds of structural heterogeneities. The approach is general and can be applied to any model transformation language. In this paper we report on its application to ATL., Work funded by the Spanish Ministry of Science (projects TIN2008-02081 and TIN2009-11555), and the R&D programme of the Madrid Region (project S2009 /TIC-1650)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-21034-1_16, Proceedings of 8th Extended Semantic Web Conference, ESWC 2011, Heraklion, Crete, Greece, May 29-June 2, 2011, We present a two-layer OWL ontology-based Knowledge Base (KB) that allows for flexible content selection and discourse structuring in Natural Language text Generation (NLG) and discuss its use for these two tasks. The first layer of the ontology contains an application-independent base ontology. It models the domain and was not designed with NLG in mind. The second layer, which is added on top of the base ontology, models entities and events that can be inferred from the base ontology, including inferable logico-semantic relations between individuals. The nodes in the KB are weighted according to learnt models of content selection, such that a subset of them can be extracted. The extraction is done using templates that also consider semantic relations between the nodes and a simple user profile. The discourse structuring submodule maps the semantic relations to discourse relations and forms discourse units to then arrange them into a coherent discourse graph. The approach is illustrated and evaluated on a KB that models the First Spanish Football League.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-20161-5_30, Proceedings of 33rd European Conference on IR Research, ECIR 2011, Dublin, Ireland, April 18-21, 2011., Collaborative Filtering (CF) aims at predicting unknown ratings of a user from other similar users. The uniqueness of the problem has made its formulation distinctive to other information retrieval problems. While the formulation has proved to be effective in rating prediction tasks, it has limited the potential connections between these algorithms and Information Retrieval (IR) models. In this paper we propose a common notational framework for IR and rating-based CF, as well as a technique to provide CF data with a particular structure, in order to be able to use any IR weighting function with it. We argue that the flexibility of our approach may yield to much better performing algorithms. In fact, in this work we have found that IR models perform well in item ranking tasks, along with different normalization strategies., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02), and the Regional Government of Madrid (S2009TIC-1542)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-22362-4_37, Proceedings of 19th International Conference, UMAP 2011, Girona, Spain, July 11-15, 2011., Research on Recommender Systems has barely explored the issue of adapting a recommendation strategy to the user’s information available at a certain time. In this thesis, we introduce a component that allows building dynamic recommendation strategies, by reformulating the performance prediction problem in the area of Information Retrieval to that of recommender systems. More specifically, we investigate a number of adaptations of the query clarity predictor in order to infer the ambiguity in user and item profiles. The properties of each predictor are empirically studied by, first, checking the correlation of the predictor output with a performance measure, and second, by incorporating a performance predictor into a recommender system to produce a dynamic strategy. Depending on how the predictor is integrated with the system, we explore two different applications: dynamic user neighbour weighting and hybrid recommendation. The performance of such dynamic strategies is examined and compared with that of static ones., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02) and Dirección General de Universidades e Investigación de la Comunidad de Madrid and Universidad Autónoma de Madrid (CCG10-UAM/TIC-5877)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-20407-4_14, Proceedings of the 14th European Conference, EuroGP 2011, Torino, Italy, April 27-29, 2011., Many different metrics have been defined in Genetic Programming. Depending on the experiment requirements and objectives, a collection of measures are selected in order to achieve an understanding of the algorithm behaviour. One of the most common metrics is the accumulated success probability, which evaluates the probability of an algorithm to achieve a solution in a certain generation. We propose a model of accumulated success probability composed by two parts, a binomial distribution that models the total number of success, and a lognormal approximation to the generation-to-success, that models the variation of the success probability with the generation., This work was partially supported by the MICYT project ABANT (TIN2010-19872) and Castilla-La Mancha project PEII09- 0266-6640

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-23734-8_30, Proceedings of 8th International Conference, CDVE 2011, Hong Kong, China, September 11-14, 2011., In this paper we present major results of a detailed study about the functionalities that are present in different collaborative systems, realized as collaborative components. We have used this study to establish a methodology for the automatic generation of collaborative applications supporting group needs. The methodology is directed to any community of end users, who do not need to have any programming skills., This research was partly funded by the Spanish National Plan of R+D, project number TIN2008-02081/TIN and by the CAM (Autonomous Community of Madrid), project number S2009/TIC-1650.

Proceedings of Third International Conference, ICTIR 2011, Bertinoro, Italy, September 12-14, 2011., The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-23318-0_5, Performance prediction is an appealing problem in Recommender Systems, as it enables an array of strategies for deciding when to deliver or hold back recommendations based on their foreseen accuracy. The problem, however, has been barely addressed explicitly in the area. In this paper, we propose adaptations of query clarity techniques from ad-hoc Information Retrieval to define performance predictors in the context of Recommender Systems, which we refer to as user clarity. Our experiments show positive results with different user clarity models in terms of the correlation with single recommender’s performance. Empiric results show significant dependency between this correlation and the recommendation method at hand, as well as competitive results in terms of average correlation., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02), University Autónoma de Madrid and the Community of Madrid (CCG10-UAM/TIC-5877)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-16020-2_42, Proceedings of 5th European Conference on Technology Enhanced Learning, EC-TEL 2010, Barcelona, Spain, September 28 - October 1, 2010., We present an e-learning system based on online forms that allows teachers to easily organise competitions in a classroom. This system is used in a preliminary study to evaluate whether cooperative competition is positive or not in education, and to identify which are the characteristics this kind of activity should have to be no harmful for students, motivating and helping them in their learning process., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02), and the Community of Madrid (S2009TIC- 1542).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-15928-2_18, Proceedings of 5th International Conference, ICGT 2010, Enschede, The Netherlands, September 27–October 2, 2010, Graph transformation is being increasingly used to express the semantics of domain specific visual languages since its graphical nature makes rules intuitive. However, many application domains require an explicit handling of time in order to represent accurately the behaviour of the real system and to obtain useful simulation metrics. Inspired by the vast knowledge and experience accumulated by the discrete event simulation community, we propose a novel way of adding explicit time to graph transformation rules. In particular, we take the event scheduling discrete simulation world view and incorporate to the rules the ability of scheduling the occurrence of other rules in the future. Hence, our work combines standard, efficient techniques for discrete event simulation (based on the handling of a future event set) and the intuitive, visual nature of graph transformation. Moreover, we show how our formalism can be used to give semantics to other timed approaches., Work partially sponsored by the Spanish Ministry of Science and Innovation, under project “METEORIC” (TIN2008-02081) and mobility grants JC2009-00015 and PR2009-0019, as well as by the R&D programme of the Community of Madrid, project “e-Madrid” (S2009/TIC-1650). We are grateful to the anonymous reviewers, which helped in improving the paper

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12127-2_11, Proceedings of 9th International Workshop, MCS 2010, Cairo, Egypt, April 7-9, 2010., This article introduces a double pruning algorithm that can be used to reduce the storage requirements, speed-up the classification process and improve the performance of parallel ensembles. A key element in the design of the algorithm is the estimation of the class label that the ensemble assigns to a given test instance by polling only a fraction of its classifiers. Instead of applying this form of dynamical (instance-based) pruning to the original ensemble, we propose to apply it to a subset of classifiers selected using standard ensemble pruning techniques. The pruned subensemble is built by first modifying the order in which classifiers are aggregated in the ensemble and then selecting the first classifiers in the ordered sequence. Experiments in benchmark problems illustrate the improvements that can be obtained with this technique. Specifically, using a bagging ensemble of 101 CART trees as a starting point, only the 21 trees of the pruned ordered ensemble need to be stored in memory. Depending on the classification task, on average, only 5 to 12 of these 21 classifiers are queried to compute the predictions. The generalization performance achieved by this double pruning algorithm is similar to pruned ordered bagging and significantly better than standard bagging.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12433-4_45, Proceedings of 8th International Conference on Practical Applications of Agents and Multiagent Systems, Regular expressions, or simply regex, have been widely used as a powerful pattern matching and text extractor tool through decades. Although they provide a powerful and flexible notation to define and retrieve patterns from text, the syntax and the grammatical rules of these regex notations are not easy to use, and even to understand. Any regex can be represented as a Deterministic or Non-Deterministic Finite Automata; so it is possible to design a representation to automatically build a regex, and a optimization algorithm able to find the best regex in terms of complexity. This paper introduces both, a graph-based representation for regex, and a particular heuristic-based evolutionary computing algorithm based on grammatical features from this language in a particular data extraction problem., This work has been partially supported by the Spanish Ministry of Science and Innovation under the projects Castilla-La Mancha project PEII09-0266-6640, COMPUBIODIVE (TIN2007-65989), and by HADA (TIN2007-64718).

Proceedings of 8th International Conference on Practical Applications of Agents and Multiagent Systems, held in Salamanca (Spain), on April 26-28, 2010, The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12433-4_50, Confidence measures play a very important role in keyword spotting systems. Traditional confidence measures are based on the score computed when the audio is decoded. Classification-based techniques by means of Multi-layer Perceptrons (MLPs) and Support Vector Machines have shown to be powerful ways to improve the final performance in terms of hits and false alarms. In this work we evaluate a keyword spotting system performance by incorporating an evolutionary algorithm as confidence measure and compare its performance with traditional classification techniques based on MLP. We show that this evolutionary algorithm gets better performance than the MLP when False Alarm (FA) is high and always performs better than the confidence measure based on the single score computed during the audio decoding., The authors acknowledge support from the Spanish Ministerio de Ciencia e innovacin, project TIN2007-66862-C02-02, and from CAM/UAM, project number CCG08-UAM/TIC-4428.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12397-9_16, Revised Selected Papers of Second COST 2102 International Training School, Dublin, Ireland, March 23-27, 2009, In this paper we present an autonomous biometric device developed in the framework of a national project. This system is able to capture speech, hand-geometry, online signature and face, and can open a door when the user is positively verified. Nevertheless the main purpose is to acquire a database without supervision (normal databases are collected in the presence of a supervisor that tells you what to do in front of the device, which is an unrealistic situation). This system will permit us to explain the main differences between what we call "real conditions" as opposed to "laboratory conditions"., This work has been supported by FEDER and MEC, TEC2006-13141-C03/TCM, and COST-2102

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12275-0_34, Proceedings of 32nd European Conference on IR Research, ECIR 2010, Milton Keynes, UK, March 28-31, 2010., Performance prediction has gained increasing attention in the IR field since the half of the past decade and has become an established research topic in the field. The present work restates the problem in the area of Collaborative Filtering (CF), where it has barely been researched so far. We investigate the adaptation of clarity-based query performance predictors to predict neighbor performance in CF. A predictor is proposed and introduced in a kNN CF algorithm to produce a dynamic variant where neighbor ratings are weighted based on their predicted performance. The properties of the predictor are empirically studied by, first, checking the correlation of the predictor output with a proposed measure of neighbor performance. Then, the performance of the dynamic kNN variant is examined on different sparsity and neighborhood size conditions, where the variant consistently outperforms the baseline algorithm, with increasing difference on small neighborhoods., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02) and the Ministry of Industry, Tourism and Commerce (CENIT-2007-1012)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-16145-2_2, Proceedings of 13th International Conference, MODELS 2010, Oslo, Norway, October 3-8, 2010., Meta-modelling is a key technique in Model Driven Engineering, where it is used for language engineering and domain modelling. However, mainstream approaches like the OMG’s Meta-Object Facility provide little support for abstraction, modularity, reusability and extendibility of (meta-)models, behaviours and transformations. In order to alleviate this weakness, we bring three elements of generic programming into meta-modelling: concepts, templates and mixin layers. Concepts permit an additional typing for models, enabling the definition of behaviours and transformations independently of meta-models, making specifications reusable. Templates use concepts to express requirements on their generic parameters, and are applicable to models and meta-models. Finally, we define functional layers by means of meta-model mixins which can extend other meta-models. As a proof of concept we also report on MetaDepth, a multi-level meta-modelling framework that implements these ideas., Work sponsored by the Spanish Ministry of Science, project TIN2008-02081 and mobility grants JC2009-00015 and PR2009-0019, and by the R&D programme of the Community of Madrid, project S2009/TIC-1650

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-12275-0_37, Proceedings of 32nd European Conference on IR Research, ECIR 2010, Milton Keynes, UK, March 28-31, 2010., Web search personalization aims to adapt search results to a user based on his tastes, interests and needs. The way in which such personal preferences are captured, modeled and exploited distinguishes the different personalization strategies. In this paper, we propose to represent a user profile in terms of social tags, manually provided by users in folksonomy systems to describe, categorize and organize items of interest, and investigate a number of novel techniques that exploit the users’ social tags to re-rank results obtained with a Web search engine. An evaluation conducted with a dataset from Delicious social bookmarking system shows that our personalization techniques clearly outperform state of the art approaches., This research was supported by the European Commission under contract FP6-027122-SALERO and by the Spanish Ministry of Science and Education (TIN2008-06566-C04-02)

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-13688-7_4, Proceedings of Third International Conference, ICMT 2010, Malaga, Spain, June 28-July 2, 2010, Graph transformation (GT) is being increasingly used in Model Driven Engineering (MDE) to describe in-place transformations like animations and refactorings. For its practical use, rules are often complemented with OCL application conditions. The advancement of rule post-conditions into pre-conditions is a well-known problem in GT, but current techniques do not consider OCL. In this paper we provide an approach to advance post-conditions with arbitrary OCL expressions into pre-conditions. This presents benefits for the practical use of GT in MDE, as it allows: (i) to automatically derive pre-conditions from the meta-model integrity constraints, ensuring rule correctness, (ii) to derive pre-conditions from graph constraints with OCL expressions and (iii) to check applicability of rule sequences with OCL conditions., Work funded by the Spanish Ministry of Science and Innovation through projects “Design and construction of a Conceptual Modeling Assistant” (TIN2008-00444/TIN - Grupo Consolidado), “METEORIC” (TIN2008-02081), mobility grants JC2009-00015 and PR2009-0019, and the R&D program of the Community of Madrid (S2009/TIC-1650, project “e-Madrid”). We would like to thank the anonymous referees and the members of the Conceptual Modeling Group (GMC) at UPC for their useful comments.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-01793-3_50, Proceedings of Third International Conference, ICB 2009, Alghero, Italy, In this paper we propose the use of Support Vector Machine Regression (SVR) for robust speaker verification in two scenarios: i) strong mismatch in speech conditions and ii) forensic environment. The proposed approach seeks robustness to situations where a proper background database is reduced or not present, a situation typical in forensic cases which has been called database mismatch. For the mismatching condition scenario, we use the NIST SRE 2008 core task as a highly variable environment, but with a mostly representative background set coming from past NIST evaluations. For the forensic scenario, we use the Ahumada III database, a public corpus in Spanish coming from real authored forensic cases collected by Spanish Guardia Civil. We show experiments illustrating the robustness of a SVR scheme using a GLDS kernel under strong session variability, even when no session variability is applied, and especially in the forensic scenario, under database mismatch., This work has been supported by the Spanish Ministry of Education under project TEC2006-13170-C02-01

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-02408-5_7, Proceedings of Second International Conference, ICMT 2009, Zurich, Switzerland, June 29-30, 2009, Pattern-based model-to-model transformation is a new approach for specifying transformations in a declarative, relational and formal style. The language relies on patterns describing allowed or forbidden relations between two models, which are compiled into operational mechanisms to perform forward and backward transformations. In this paper, we extend the approach for handling attribute conditions expressed in some suitable logic, adapt the operational mechanisms based on graph transformation to relax attribute handling by constraint solving, and discuss heuristics for the compilation of patterns into rules., Work supported by the Spanish Ministry of Science and Innovation, projects METEORIC (TIN2008-02081),MODUWEB (TIN2006-09678) and FORMALISM (TIN2007-66523).Moreover, part of this work was done during a sabbatical leave of the third author at TU Berlin, with financial support from the Ministerio de Ciencia e Innovaci´on (grant ref. PR2008-0185). We thank the referees for their useful comments

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-02481-8_117, Proceedings of 10th International Work-Conference on Artificial Neural Networks, IWANN 2009 Workshops, Salamanca, Spain, June 10-12, 2009, This paper covers those aspects of modern interfaces which expand and enhance the way in which people interact with computers, like multi-touch table systems, presence-detection led displays and interactive virtualized real-life environments. It elaborates on how disabled or conditioned people take great advantage of natural interaction as interfaces adapt to their needs; interfaces which can be focused towards memory, cognitive or physical deficiencies. Applications size-up to serve specific users with customized tools and options, and are aware while taking into account the state and situation of the individual., This work has been partly funded by HADA project number TIN2007 – 64718.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-04274-4_10, Proceedings of 19th International Conference, Limassol, Cyprus, September 14-17, 2009., Recent research has shown that the provisional count of votes of an ensemble of classifiers can be used to estimate the probability that the final ensemble prediction coincides with the current majority class. For a given instance, querying can be stopped when this probability is above a specified threshold. This instance-based ensemble pruning procedure can be efficiently implemented if these probabilities are pre-computed and stored in a lookup table. However, the size of the table and the cost of computing the probabilities grow very rapidly with the number of classes of the problem. In this article we introduce a number of computational optimizations that can be used to make the construction of the lookup table feasible. As a result, the application of instance-based ensemble pruning is extended to multi-class problems. Experiments in several UCI multi-class problems show that instance-based pruning speeds-up classification by a factor between 2 and 10 without any significant variation in the prediction accuracy of the ensemble.

Proceedings of 8th International Conference, FQAS 2009, Roskilde, Denmark, October 26-28, 2009., The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-04957-6_52, Performance prediction has gained increasing attention in the IR field since the half of the past decade and has become an established research topic in the field. The present work restates the problem in the subarea of Collaborative Filtering (CF), where it has barely been researched so far. We investigate the adaptation of clarity-based query performance predictors to define predictors of neighbor performance in CF. The proposed predictors are introduced in a memory-based CF algorithm to produce a dynamic variant where neighbor ratings are weighted based on their predicted performance. The approach is tested with encouraging empirical results, as the dynamic variants consistently outperform the baseline algorithms, with increasing difference on small neighborhoods., This work was supported by the Spanish Ministry of Science and Innovation (TIN2008-06566-C04-02) and the Ministry of Industry, Tourism and Commerce (CENIT-2007-1012).

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-02478-8_59, Proceedings of 10th International Work-Conference on Artificial Neural Networks, IWANN 2009, Salamanca, Spain., This work tests the suitability of NEPs to parse languages. We propose PNEP, a simple extension to NEP, and a procedure to translate a grammar into a PNEP that recognizes the same language. These parsers based on NEPs do not impose any additional constrain to the structure of the grammar, which can contain all kinds of recursive, lambda or ambiguous rules. This flexibility makes this procedure specially suited for Natural Languge Processing (NLP). In a first proof with a simplified English grammar, we got a performance (a linear time complexity) similar to that of the most popular syntactic parsers in the NLP area (Early and its derivatives). All the possible derivations for ambiguous grammars were generated, This work was partially supported by MEC, project TIN2008-02081/TIN and by DGUI CAM/UAM, project CCG08-UAM/TIC-4425.

Revised Selected Papers of 4th International Conference, ICSOFT 2009, Sofia, Bulgaria, July 26-29, 2009., The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-20116-5_17, Scheduling a set of constrained resources is a difficult task, specially when there is no clear definition of ‘optimal’. When the constraints depend not only on physical or temporal issues but also in human desires or preferences the task gets harder. This is the case of educational resources, for example when a set of students must be distributed into a limited set of laboratories to attend to periodical practical sessions, in this case weekly. The preferences of the students may vary during the process for reasons such as the number of people already in that group. This paper presents a socio–inspired solution implemented as a multiagent system. The agents enroll themselves in the lab sessions based on their preferences and negotiate with other agents, using the resources they already have, to obtain desired groups that were already full., This work has been partially supported by the Spanish Ministry of Science and Innovation under grants TIN 2007-65989, TIN 2007-64718, and TIN2010-19872.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-642-03741-2_26, Proceedings of Third International Conference, CALCO 2009, Udine, Italy, September 7-10, 2009., Model-to-model (M2M) transformation consists in trans- forming models from a source to a target language. Many transformation languages exist, but few of them combine a declarative and relational style with a formal underpinning able to show properties of the transformation. Pattern-based transformation is an algebraic, bidirectional, and relational approach to M2M transformation. Specifications are made of patterns stating the allowed or forbidden relations between source and target models, and then compiled into low level operational mechanisms to perform source-to-target or target-to-source transformations. In this paper, we study the compilation into operational triple graph grammar rules and show: (i) correctness of the compilation of a specification without negative patterns; (ii) termination of the rules, and (iii) completeness, in the sense that every model considered relevant can be built by the rules., Work supported by the Spanish Ministry of Science and Innovation, projects METEORIC (TIN2008-02081), MODUWEB (TIN2006-09678) and FORMALISM (TIN2007-66523). Moreover, part of this work was done during a sabbatical leave of the first author at TU Berlin, with financial support from the Spanish Ministry of Science and Innovation (grant ref. PR2008-0185). We thank the referees for their useful comments