Search

Breast cancer is one of the most dangerous diseases that attack women in their 40s worldwide. Due to this fact, it is estimated that one in eight women will develop a malignant carcinoma during their life. In addition, the carelessness of performing regular screenings is an important reason for the increase of mortality. However, computer-aided diagnosis systems attempt to enhance the quality of mammograms as well as the detection of early signs related to the disease. In this paper we propose a bank of Gabor filters to calculate the mean, standard deviation, skewness and kurtosis features by four-sized evaluation windows. Therefore, an active strategy is used to select the most relevant pixels. Finally, a supervised classification stage using two-class support vector machines is utilised through an accurate estimation of kernel parameters. In order to show the development of our methodology based on mammographic image analysis, two main experiments are fulfilled: abnormal/normal breast tissue classificat...

Global change refers to climate changes, but also demographic, technological and economic changes. Predicted water scarcity will be critical in the coastal Mediterranean region, especially for provision to mid-sized and large-sized cities. This paper studies the case of the city of Tarragona, located at the Mediterranean area of north-eastern Spain (Catalonia). Several scenarios have been constructed to evaluate different sectorial water allocation policies to mitigate the water scarcity induced by global change. Future water supply and demand predictions have been made for three time spans. The decision support system presented is based on the outranking model, which constructs a partial pre-order based on pairwise preference relations among all the possible actions. The system analyses a hierarchical structure of criteria, including environmental and economic criteria. We compare several adaptation measures including alternative water sources, inter-basin water transfer and sectorial demand management c...

Background: The aim of this study was to build a clinical decision support system (CDSS) in diabetic retinopathy (DR), based on type 2 diabetes mellitus (DM) patients. Method: We built a CDSS from a sample of 2,323 patients, divided into a training set of 1,212 patients, and a testing set of 1,111 patients. The CDSS is based on a fuzzy random forest, which is a set of fuzzy decision trees. A fuzzy decision tree is a hierarchical data structure that classifies a patient into several classes to some level, depending on the values that the patient presents in the attributes related to the DR risk factors. Each node of the tree is an attribute, and each branch of the node is related to a possible value of the attribute. The leaves of the tree link the patient to a particular class (DR, no DR). Results: A CDSS was built with 200 trees in the forest and three variables at each node. Accuracy of the CDSS was 80.76%, sensitivity was 80.67%, and specificity was 85.96%. Applied variables were current age, gender, DM duration and treatment, arterial hypertension, body mass index, HbA1c, estimated glomerular filtration rate, and microalbuminuria. Discussion: Some studies concluded that screening every 3 years was cost effective, but did not personalize risk factors. In this study, the random forest test using fuzzy rules permit us to build a personalized CDSS. Conclusions: We have developed a CDSS that can help in screening diabetic retinopathy programs, despite our results more testing is essential.

Multi-Criteria Decision Aid methods are used to find the best option from a set of alternatives when multiple and conflicting criteria have to be optimized simultaneously. The evaluation of the suitability or risk of each alternative is usually performed by assigning a numerical value. However, sometimes the data required to measure a criterion may be found in the form of semantic values such as tags. This paper proposes a methodology to calculate the strength of an outranking relation for a pair of alternatives using semantic criteria following the principles of ELECTRE-III (i.e. by means of concordance and discordance indices). The preferences about semantic data are represented in an ontology by means of objective and subjective functions. The paper explains how this new methodology was applied to analyse different electricity generation technologies using environmental and economic criteria. Two scenarios are tested to show how semantic criteria may influence the final decision.

Comunicación presentada al 15th International Conference on Modeling Decisions for Artificial Intelligence, MDAI 2018 (15 - 18 october 2018). Fuzzy measures are used to express background knowledge of the information sources. In fuzzy rule-based models, the rule confidence gives an important information about the final classes and their relevance. This work proposes to use fuzzy measures and integrals to combine rules confidences when making a decision. A Sugeno $$\lambda $$ -measure and a distorted probability have been used in this process. A clinical decision support system (CDSS) has been built by applying this approach to a medical dataset. Then we use our system to estimate the risk of developing diabetic retinopathy. We show performance results comparing our system with others in the literature. This work is supported by the URV grant 2017PFR-URV-B2-60, and by the Spanish research projects no: PI12/01535 and PI15/01150 for (Instituto de Salud Carlos III and FEDER funds). Mr. Saleh has a Pre-doctoral grant (FI 2017) provided by the Catalan government and an Erasmus+ travel grant by URV. Prof. Bustince acknowledges the support of Spanish project TIN2016-77356-P.

Some decision aiding methods are based on constructing and exploiting outranking relations. An alternative a outranks another b if a is at least as good as b (aSb). One well known method in this field is ELECTRE. The outranking relation is usually built by means of a weighted average (WA) of the votes given by a set of criterion with respect to the fulfilment of aSb. The value obtained represent the strength of the majority opinion. The WA operator can be observed to have sometimes an undesired compensative effect. In this paper we propose the use of other aggregation operators with different mathematical properties. In particular, we substitute the WA by three operators from the Ordered Weighted Average (OWA) family of operators because it permits to decide the degree of andness/orness that is used during the aggregation. The OWAWA (Ordered Weighted Average Weighted Average), WOWA (Weighted Ordered Weighted Average) and IOWA (Induced Ordered Weighted Average) operators are studied. They are capable to combine the importance given to each criterion with the conjunctive/disjunctive requirement applied in the definition of the outranking relation.

Intelligent data analysis methods usually require as input a matrix, in which each row is an object to be analysed and each column is an attribute. In most cases it is assumed that attributes are Boolean, categorical or numerical. With the advent of semantic domain information in the form of ontologies, it is now common to find also semantic attributes, which may take as value a list of concepts. This paper proposes a new ontology-based procedure to compute the similarity between lists of semantic values, which may be used to compare objects. This measure is employed in an enhanced version of the k-means clustering method. The usefulness of the obtained classes has been tested in the context of a Web-based personalised recommender of Tourist destinations.

10.1007/s10115-013-0663-5 Today, it is well known that taking into account the semantic information available for categorical variables sensibly improves the meaningfulness of the final results of any analysis. The paper presents a generalization of mixed Gibert¿s metrics, which originally handled numerical and categorical variables, to include also semantic variables. Semantic variables are defined as categorical variables related to a reference ontology (ontologies are formal structures to model semantic relationships between the concepts of a certain domain). The superconcept-based distance (SCD) is introduced to compare semantic variables taking into account the information provided by the reference ontology. A benchmark shows the good performance of SCD with respect to other proposals, taken from the literature, to compare semantic features. Mixed Gibert¿s metrics is generalized incorporating SCD. Finally, two real applications based on touristic data show the impact of the generalized Gibert¿s metrics in clustering procedures and, in consequence, the impact of taking into account the reference ontology in clustering. The main conclusion is that the reference ontology, when available, can sensibly improve the meaningfulness of the final clusters.

Diabetic retinopathy is one of the most common comorbidities of diabetes. Unfortunately, the recommended annual screening of the eye fundus of diabetic patients is too resource-consuming. Therefore, it is necessary to develop tools that may help doctors to determine the risk of each patient to attain this condition, so that patients with a low risk may be screened less frequently and the use of resources can be improved. This paper explores the use of two kinds of ensemble classifiers learned from data: fuzzy random forest and dominance-based rough set balanced rule ensemble. These classifiers use a small set of attributes which represent main risk factors to determine whether a patient is in risk of developing diabetic retinopathy. The levels of specificity and sensitivity obtained in the presented study are over 80%. This study is thus a first successful step towards the construction of a personalized decision support system that could help physicians in daily clinical practice.

The final publication is available at Springer via http://dx.doi.org/10.1007/s10479-017-2590-4 This paper considers the problem of finding suitable sites for wind farms in a region of Catalonia (Spain). The evaluation criteria are structured into a hierarchy that identifies several intermediate sub-goals dealing with different points of view. Therefore, the recent ELECTRE-III-H hierarchical multi-criteria analysis method is proposed as a good solution to help decision-makers. This method establishes an order among the set of possible sites for the wind farms for each sub-goal. ELECTRE-III-H aggregates these orders into an overall order using different parameters. The procedure is based on the construction and exploitation of a pairwise outranking relation, following the principles of concordance (i.e. majority rule) and discordance (i.e. respect for the minority opinions). This paper makes two main contributions. First, it contributes to the ELECTRE-III-H method by studying its mathematical properties for the construction of outranking relations. Second, the case study is solved and its results show that we can effectively represent and manage the overall influence of the various criteria on the global result at different levels of the hierarchy. The paper compares different scenarios with strict, normal, and optimistic preference, indifference and veto thresholds. Results show that the best site differs for technical, economic, environmental, and social intermediate criteria. Therefore, the best overall solution changes depending on the preference and veto thresholds fixed at the intermediate level of the hierarchy.

Recommender systems are very useful in domains in which a large amount of continuous information needs to be evaluated before a decision is made. Systems that permanently interact with users need to be adapted to changes in their interests. This paper proposes an algorithm that takes advantage of the preference information implicit in the actions of the user to dynamically adapt the user profile, in which user preferences are represented as fuzzy sets. The algorithm has been tested with real data extracted from the New York Times and has shown promising results. This paper presents the adaptation algorithm and discusses the influence of its basic parameters.

SigTur/E-Destination is a Web-based system that provides personalized recommendations of touristic activities in the region of Tarragona. The activities are properly classified and labeled according to a specific ontology, which guides the reasoning process. The recommender takes into account many different kinds of data: demographic information, travel motivations, the actions of the user on the system, the ratings provided by the user, the opinions of users with similar demographic characteristics or similar tastes, etc. The system has been fully designed and implemented in the Science and Technology Park of Tourism and Leisure. The paper presents a numerical evaluation of the correlation between the recommendations and the user's motivations, and a qualitative evaluation performed by end users.

Using microdata provided by statistical agencies has many benefits from the data mining point of view. However, such data often involve sensitive information that can be directly or indirectly related to individuals. An appropriate anonymisation process is needed to minimise the risk of disclosure. Several masking methods have been developed to deal with continuous-scale numerical data or bounded textual values but approaches to tackling the anonymisation of textual values are scarce and shallow. Because of the importance of textual data in the Information Society, in this paper we present a new masking method for anonymising unbounded textual values based on the fusion of records with similar values to form groups of indistinguishable individuals. Since, from the data exploitation point of view, the utility of textual information is closely related to the preservation of its meaning, our method relies on the structured knowledge representation given by ontologies. This domain knowledge is used to guide the masking process towards the merging that best preserves the semantics of the original data. Because textual data typically consist of large and heterogeneous value sets, our method provides a computationally efficient algorithm by relying on several heuristics rather than exhaustive searches. The method is evaluated with real data in a concrete data mining application that involves solving a clustering problem. We also compare the method with more classical approaches that focus on optimising the value distribution of the dataset. Results show that a semantically grounded anonymisation best preserves the utility of data in both the theoretical and the practical setting, and reduces the probability of record linkage. At the same time, it achieves good scalability with regard to the size of input data.

Estimation of the semantic likeness between words is of great importance in many applications dealing with textual data such as natural language processing, knowledge acquisition and information retrieval. Semantic similarity measures exploit knowledge sources as the base to perform the estimations. In recent years, ontologies have grown in interest thanks to global initiatives such as the Semantic Web, offering an structured knowledge representation. Thanks to the possibilities that ontologies enable regarding semantic interpretation of terms many ontology-based similarity measures have been developed. According to the principle in which those measures base the similarity assessment and the way in which ontologies are exploited or complemented with other sources several families of measures can be identified. In this paper, we survey and classify most of the ontology-based approaches developed in order to evaluate their advantages and limitations and compare their expected performance both from theoretical and practical points of view. We also present a new ontology-based measure relying on the exploitation of taxonomical features. The evaluation and comparison of our approach's results against those reported by related works under a common framework suggest that our measure provides a high accuracy without some of the limitations observed in other works.

In the context of Statistical Disclosure Control, microaggregation is a privacy-preserving method aimed to mask sensitive microdata prior to publication. It iteratively creates clusters of, at least, k elements, and replaces them by their prototype so that they become k-indistinguishable (anonymous). This data transformation produces a loss of information with regards to the original dataset which affects the utility of masked data, so, the aim of microaggregation algorithms is to find the partition that minimises the information loss while ensuring a certain level of privacy. Most microaggregation methods, such as the MDAV algorithm, which is the focus of this paper, have been designed for numerical data. Extending them to support non-numerical (categorical) attributes is not straightforward because of the limitations on defining appropriate aggregation operators. Concretely, related works focused on the MDAV algorithm propose grouping data into groups with constrained size (or even fixed) and/or incorporate a basic categorical treatment of non-numerical data. This approach affects negatively the utility of the protected dataset because neither the distributional characteristics of data nor their underlying semantics are properly considered. In this paper, we propose a set of modifications to the MDAV algorithm focused on categorical microdata. Our approach has been evaluated and compared with related works when protecting real datasets with textual attribute values. Results show that our method produces masked datasets that better minimises the information loss resulting from the data transformation.

Recommender systems in e-Tourism normally focus on helping tourists to select appropriate destinations. A related problem that has been less explored in the literature is how to provide personalised recommendations of cultural and leisure activities when the tourist has already arrived at the destination. This paper presents a novel recommendation system, [email protected], which addresses this issue. Its agent-based modular design permits to model different kinds of activities in a flexible way, and allows the implementation of a location-aware front-end in the mobile device of the user. Special care has been put in the recommendation engine, implemented via a specialised Recommender Agent. It incorporates a mixture of content-based and collaborative recommendation strategies, thus avoiding the drawbacks of each individual method, and is able to perform recommendations in heterogeneous scenarios. Recommendations take into account user profiles which are implicitly updated after the analysis of user actions (e.g., queries, evaluations). The system has been successfully deployed and tested in the World Heritage-listed city of Tarragona.

In complex domains it is usually quite difficult to introduce context information. However, sometimes that information should be taken into account to make decisions, because it provides some relevant knowledge that cannot be expressed using an attribute-value representation. This is the case of the determination of risk of contamination of soils. In this paper, we propose to use conjunctive rules to introduce additional background knowledge to a MCDM sorting method called ClusDM. ClusDM is based on the aggregation of the data with unsupervised clustering techniques. The paper presents a new algorithm to incorporate rules to guide the clustering process in a semi-supervised way. The paper also describes how it works in the case sorting a set of possible contaminated soils, and compares the results obtained by ClusDM when rules are used or not.

Background: Prospective, population-based study of an 8-year follow up. To determine the direct cost of diabetic retinopathy [DR], evaluating our screening programme and the cost of treating DR, focusing on diabetic macular oedema [DMO] after anti-vascular endothelial growth factor [anti-VEGF] treatment. Methods: A total of 15,396 diabetes mellitus [DM] patients were studied. We determined the cost-effectiveness of our screening programme against an annual programme by applying the Markov simulation model. We also compared the cost-effectiveness of anti-VEGF treatment to laser treatment for screened patients with DMO. Results: The cost of our 2.5-year screening programme was as follows: per patient with any-DR, €482.85 ± 35.14; per sight-threatening diabetic retinopathy [STDR] patient, €1528.26 ± 114.94; and €1826.98 ± 108.26 per DMO patient. Comparatively, an annual screening programme would result in increases as follows: 0.77 in QALY per patient with any-DR and 0.6 and 0.44 per patient with STDR or DMO, respectively, with an incremental cost-effective ratio [ICER] of €1096.88 for any-DR, €4571.2 for STDR and €7443.28 per DMO patient. Regarding diagnosis and treatment, the mean annual total cost per patient with DMO was €777.09 ± 49.45 for the laser treated group and €7153.62 ± 212.15 for the anti-VEGF group, with a QALY gain of 0.21, the yearly mean cost was €7153.62 ± 212.15 per patient, and the ICER was €30,361. Conclusions: Screening for diabetic retinopathy every 2.5 years is cost-effective, but should be adjusted to a patient's personal risk factors. Treatment with anti-VEGF for DMO has increased costs, but the cost-utility increases to 0.21 QALY per patient.

Erratum

ELECTRE-III is a well-known multi-criteria decision analysis method that ranks a set of alternatives in terms of a set of heterogeneous evaluation criteria. It is based on constructing and exploiting a pairwise outranking relation between alternatives, which are defined with numerical and ordinal values. However, nowadays it is very common the use of descriptive linguistic tags, which is information that requires a qualitative treatment rather than a numerical one. In this paper we propose to store the user preferences about a set of tags in an ontological structure and to use this knowledge to construct the outranking relation by means of a semantic analysis of the tags associated to the alternatives. Uncertainty is handled by means of fuzzy concordance and discordance functions. The method is illustrated with a case study related to the recommendation of touristic activities.

Diabetic retinopathy is an ocular disease that involves an important healthcare spending and is the most serious cause of secondary blindness. Precocious and precautionary detection through a yearly screening of the eye fundus is difficult to make because of the large number of diabetic patients. This paper presents a novel clinical decision support system, based on fuzzy rules, that calculates the risk of developing diabetic retinopathy. The system has been trained and validated on a dataset of patients from Sant Joan de Reus University Hospital. The system achieves levels of sensitivity and specificity above 80 %, which is in practice the minimum threshold required for the validity of clinical tests.

The regional water allocation planning is one of those complex decision problems where holistic approach to water supply management considering different criteria would be valuable. However, multi-criteria decision making with diverse indicators measured on different scales and uncertainty levels is difficult to solve. Objective of this paper is to develop scenarios for the future imbalances in water supply and demand for a water stressed Mediterranean area of Northern Spain (Tarragona) and to test the applicability and suitability of an outranking method ELECTRE-III-H for evaluating sectoral water allocation policies. This study is focused on the use of alternative water supply scenarios to fulfil the demand of water from three major sectors: domestic, industrial and agricultural. A detail scenario planning for regional water demand and supply has been discussed. For each future scenario of climate change, the goal is to obtain a ranking of a set of possible actions with regards to different types of indicators (costs, water stress and environmental impact). The analytical method used is based on outranking models for decision aid with hierarchical structures of criteria and ranking alternatives using partial preorders based on pairwise preference relations. We compare several adaptation measures including alternative water sources (reclaimed water and desalination); inter basin water transfer and sectoral demand management coming from industry, agriculture and domestic sectors and tested the sustainability of management actions for different climate change scenarios. Results have shown use of alternative water resources as the most reliable alternative with medium reclaimed water reuse in industry and agriculture and low to medium use of desalination water in domestic and industrial sectors as the best alternative. The proposed method has several advantages such as the management of heterogeneous scales of measurement without requiring any artificial transformation and the management of uncertainty by means of comparisons at a qualitative level in terms of the decision maker preferences.

The Multiagent Systems Group (GruSMA) is part of the Research Group on Artificial Intelligence at Spain's University Rovira i Virgili. GruSMA specializes in agent technology, working in application areas such as mobile devices, transportation and tourism user services, automated ontology construction, and agent-based simulation of complex systems. Over the past five years, GruSMA has designed and developed many agent applications to address healthcare problems. Some of these systems are academic, proof-of-concept exercises, but they point to the kind of problems in medicine that intelligent agents can help solve in the near term

The exploitation of microdata compiled by statistical agencies is of great interest for the data mining community. However, such data often include sensitive information that can be directly or indirectly related to individuals. Hence, an appropriate anonymisation process is needed to minimise the risk of disclosing identities and/or confidential data. In the past, many anonymisation methods have been developed to deal with numerical data, but approaches tackling the anonymisation of non-numerical values (e.g. categorical, textual) are scarce and shallow. Since the utility of this kind of information is closely related to the preservation of its meaning, in this work, the notion of semantic similarity is used to enable a semantically coherent anonymisation. The knowledge modelled in ontologies is used as the basic pillar to propose semantic operators that enable an accurate management and transformation of categorical attributes. These operators are then used in three anonymisation mechanisms: Semantic Recoding, Semantic and Adaptive Microaggregation and Semantic Resampling. The three algorithms are compared in terms of semantic utility, privacy disclosure risk and runtime, with encouraging results.

Recommender systems are currently being applied in many different domains. This paper focuses on their application in tourism. A comprehensive and thorough search of the smart e-Tourism recommenders reported in the Artificial Intelligence journals and conferences since 2008 has been made. The paper provides a detailed and up-to-date survey of the field, considering the different kinds of interfaces, the diversity of recommendation algorithms, the functionalities offered by these systems and their use of Artificial Intelligence techniques. The survey also provides some guidelines for the construction of tourism recommenders and outlines the most promising areas of work in the field for the next years.

The estimation of semantic similarity between words is an important task in many language related applications. In the past, several approaches to assess similarity by evaluating the knowledge modelled in an ontology have been proposed. However, in many domains, knowledge is dispersed through several partial and/or overlapping ontologies. Because most previous works on semantic similarity only support a unique input ontology, we propose a method to enable similarity estimation across multiple ontologies. Our method identifies different cases according to which ontology/ies input terms belong. We propose several heuristics to deal with each case, aiming to solve missing values, when partial knowledge is available, and to capture the strongest semantic evidence that results in the most accurate similarity assessment, when dealing with overlapping knowledge. We evaluate and compare our method using several general purpose and biomedical benchmarks of word pairs whose similarity has been assessed by human experts, and several general purpose (WordNet) and biomedical ontologies (SNOMED CT and MeSH). Results show that our method is able to improve the accuracy of similarity estimation in comparison to single ontology approaches and against state of the art related works in multi-ontology similarity assessment.

Home Care (HC) assistance is emerging as an effective and efficient alternative to institutionalized care, especially for the case of senior patients that present multiple co-morbidities and require life long treatments under continuous supervision. The care of such patients requires the definition of specially tailored treatments and their delivery involves the coordination of a team of professionals from different institutions, requiring the management of many kinds of knowledge (medical, organizational, social and procedural). The K4Care project aims to assist the HC of elderly patients by proposing a standard HC model and implementing it in a knowledge-driven e-health platform aimed to support the provision of HC services. This paper focuses on two knowledge-based personalization aspects incorporated in the platform that aim to overcome the difficulties of HC delivery. The first one is the assistance to medical practitioners in the process of defining a customized treatment adjusted to the medical and social conditions of a particular patient in order to consider multiple co-morbidities. The second one is the possibility of tailoring the profiles of the care professionals according to the medical and organizational daily requirements in order to allow a flexible care delivery. Those two aspects, guided by the knowledge explicitly represented in the platform, play a crucial role in the medical and social acceptance of this kind of e-health systems. The paper also includes a real case study designed and tested by healthcare professionals and includes encouraging results from the test of the platform in a real health care environment in the city of Pollenza (Italy).

Privacy should be carefully considered during the publication of data (e.g. database records) collected from individuals to avoid disclosing identities or revealing confidential information. Anonymisation methods aim at achieving a certain degree of privacy by performing transformations over non-anonymous data while minimising, as much as possible, the distortion (i.e. information loss) derived from these transformations. k-anonymity is a property typically considered when masking data, stating that each record (corresponding to an individual) is indistinguishable from at least k-1 other records in the anonymised dataset. Many methods have been developed to anonymise data, but most of them are focused solely on numerical attributes. Non-numerical values (e.g. categorical attributes like job or country-of-birth or unbounded textual ones like user preferences) are more challenging because arithmetic operations cannot be applied. To properly manage and interpret this kind of data, it is required to have operators that are able to deal with data semantics. In this paper, we propose an anonymisation method based on a classic data re-sampling algorithm that guarantees the fulfilment of the k-anonymity property and is able to deal with non-numerical data from a semantic perspective. Our method has been applied to anonymise the well-known Adult Census dataset, showing that a semantic interpretation of non-numerical values better minimises the information loss of the masked data file.

Geospatial information is commonly used in tourism to facilitate activity planning, especially in a context of limited information on the territory, as it is common in the case of complex and heterogeneous tourism destination regions where the constrained spatial activity of visitor is likely to generate inefficiencies in the use of assets and resources, and hinder visitor satisfaction. Because of the large amount of spatial and non-spatial data associated with different resources and activities, it is a logical choice to use geographic information systems (GIS) for storing, managing, analyzing, and visualizing the data. Nevertheless, in order to facilitate personalized recommendations to visitors, interaction with Artificial Intelligence is needed. This chapter presents SigTur/E-Destination, a tourism recommender system based on a semantically-enriched GIS that provides regional tourist organizations and the industry with a new powerful tool for the sustainable management of their destinations. The recommendation system uses innovative Artificial Intelligence techniques, such as a hybrid method that integrates content-based and collaborative filtering and clustering methodologies that improve computational time.

Centroids are key components in many data analysis algorithms such as clustering or microaggregation. They are considered as the central value that minimises the distance to all the objects in a dataset or cluster. Methods for centroid construction are mainly devoted to datasets with numerical and categorical attributes, focusing on the numerical and distributional properties of data. Textual attributes, on the contrary, consist of term lists referring to concepts with a specific semantic content (i.e., meaning), which cannot be evaluated by means of classical numerical operators. Hence, the centroid of a dataset with textual attributes should be the term that minimises the semantic distance against the members of the set. Semantically-grounded methods aiming to construct centroids for datasets with textual attributes are scarce and, as it will be discussed in this paper, they are hampered by their limited semantic analysis of data. In this paper, we propose a method that, exploiting the knowledge provided by background ontologies (like WordNet), is able to construct the centroid of multivariate datasets described by means of textual attributes. Special efforts have been put in the minimisation of the semantic distance between the centroid and the input data. As a result, our method is able to provide optimal centroids (i.e., those that minimise the distance to all the objects in the dataset) according to the exploited background ontology and a semantic similarity measure. Our proposal has been evaluated by means of a real dataset consisting on short textual answers provided by visitors of a natural park. Results show that our centroids retain the semantic content of the input data better than related works.

In many real-world multiple criteria decision problems, the family of criteria has a hierarchical structure presented in the form of a tree. The leaves of the tree correspond to elementary criteria on which a finite set of alternatives is directly evaluated. Evaluations of alternatives on elementary criteria are aggregated to form a sub-criterion at an upper level of the tree. Then, evaluations of alternatives on sub-criteria having the same predecessor in the hierarchy tree are aggregated again in a sub-criterion of a higher level, and so on, until the aggregation at the general goal criterion, which is the root of the tree, where the alternatives are finally ranked from the best to the worst. At each node of the tree, above the leaves, we are considering the aggregation of multiple criteria evaluations using the ELECTRE-III method, based on building and exploiting outranking relations for each pair of alternatives. As the result of ELECTRE-III is a partial preorder of alternatives, the sub-criteria are ordering the alternatives just partially. Therefore, in this paper we propose a new way of calculating concordance and discordance indices that take part in the definition of an outranking relation aggregating the (partially ordered) evaluations on sub-criteria. A robustness analysis has been performed to analyze the behavior of the proposed method in different settings.

Ontologies define a set of concepts related to a certain domain as well as the relationships among them. This structure may be exploited to represent and reason about the preferences of a user. The user profile stores the degrees of interest of the user on several concepts using membership functions. In this way, each concept in the ontology is a fuzzy set and any user belongs to this fuzzy set to a certain degree. To represent the uncertainty on this information, a degree of confidence on the membership value is also included. After an initial assignment of preferences, spreading algorithms that exploit the taxonomical information of the ontology are applied to propagate the information about the user’s preferences (and their associated uncertainty) through the whole set of concepts. This framework for managing uncertain preferences has been successfully applied in a Tourism recommender system.

The development of digital and web technologies opened new horizons regarding the generation of personalized contents and the management of visitor services. The objective of SIGtur/E-Destination is to provide to the tourist and travel sector electronic tools for the sustainable management of destinations, also possibly leading to an increase in the effectiveness of visitor service firms. The SIGtur/E-Destination system includes a warehouse of geo-referenced information that updates the available information on the tourism activities in the Costa Daurada and Terres de l’Ebre in the south of Catalonia (Spain). This catalogue of information on resources, attractions, products, establishments and packages has to make possible the improvement of the interaction of between public and private agents and visitors. A recommender system uses this catalogue to offer personalized information to the tourists.

Aggregation operators for linguistic variables usually assume a uniform and symmetrical distribution of the linguistic terms that define the variable. A well-known aggregation operator is the Linguistic Ordered Weighted Average (LOWA), which has been extensively applied. However, there are some problems where an unbalanced set of linguistic terms is more appropriate to describe the objects. In this paper we define the Unbalanced Linguistic Ordered Weighted Average (ULOWA) on the basis of the LOWA operator. ULOWA takes into account the fuzzy membership functions of the terms during the aggregation process. There is no restriction on the form of the membership functions of the terms, which can be triangular or trapezoidal, non symmetrical and non equally distributed. The paper demonstrates the properties of ULOWA. Finally, a comparison of this operator with some other aggregation operators for unbalanced sets of terms is done.

Clustering was usually applied on numerical and categorical information. However, textual information is acquiring an increasing importance with the appearance of methods for textual data mining. This paper proposes the use of classical clustering algorithms with a mixed function that combines numerical, categorical and semantic features. The content of the semantic features is extracted from textual data. This paper analyses and compares the behavior of different existing semantic similarity functions that use WordNet as background ontology. The different partitions obtained with the clustering algorithm are compared to human classifications in order to see which one approximates better the human reasoning. Moreover, the interpretability of the obtained clusters is discussed. The results show that those similarity measures that provide better results when compared using a standard benchmark also provide better and more interpretable partitions.

The analysis of sensible data requires a proper anonymization of values in order to preserve the privacy of individuals. Information loss should be minimized during the masking process in order to enable a proper exploitation of data. Even though several masking methods have been designed for numerical data, very few of them deal with categorical (textual) information. In this case, the quality of the anonymized dataset is closely related to the preservation of semantics, a dimension which is commonly neglected of shallowly considered in related words. In this paper, a new masking method for unbounded categorical attributes is proposed. It relies on the knowledge modeled in ontologies in order to semantically interpret the input data and perform data transformations aiming to minimize the loss of semantic content. On the contrary to exhaustive methods based on simple hierarchical structures, our approach relies on a set of heuristics in order to guide and optimize the masking process, ensuring its scalability when dealing with big and heterogenous datasets and wide ontologies. The evaluation performed over real textual data suggests that our method is able to produce anonymized datasets which significantly preserve data semantics in comparison to apporaches based on data distribution metrics.

The version of record is available online at: http://dx.doi.org/10.1007/s10844-009-0103-x Estimation of the degree of semantic similarity/distance between concepts is a very common problem in research areas such as natural language processing, knowledge acquisition, information retrieval or data mining. In the past, many similarity measures have been proposed, exploiting explicit knowledge—such as the structure of a taxonomy—or implicit knowledge—such as information distribution. In the former case, taxonomies and/or ontologies are used to introduce additional semantics; in the latter case, frequencies of term appearances in a corpus are considered. Classical measures based on those premises suffer from some problems: in the first case, their excessive dependency of the taxonomical/ontological structure; in the second case, the lack of semantics of a pure statistical analysis of occurrences and/or the ambiguity of estimating concept statistical distribution from term appearances. Measures based on Information Content (IC) of taxonomical concepts combine both approaches. However, they heavily depend on a properly pre-tagged and disambiguated corpus according to the ontological entities in order to compute accurate concept appearance probabilities. This limits the applicability of those measures to other ontologies –like specific domain ontologies- and massive corpus –like the Web-. In this paper, several of the presented issues are analyzed. Modifications of classical similarity measures are also proposed. They are based on a contextualized and scalable version of IC computation in the Web by exploiting taxonomical knowledge. The goal is to avoid the measures’ dependency on the corpus pre-processing to achieve reliable results and minimize language ambiguity. Our proposals are able to outperform classical approaches when using the Web for estimating concept probabilities.

Proper understanding of textual data requires the exploitation and integration of unstructured and heterogeneous clinical sources, healthcare records or scientific literature, which are fundamental aspects in clinical and translational research. The determination of semantic similarity between word pairs is an important component of text understanding that enables the processing, classification and structuring of textual resources. In the past, several approaches for assessing word similarity by exploiting different knowledge sources (ontologies, thesauri, domain corpora, etc.) have been proposed. Some of these measures have been adapted to the biomedical field by incorporating domain information extracted from clinical data or from medical ontologies (such as MeSH or SNOMED CT). In this paper, these approaches are introduced and analyzed in order to determine their advantages and limitations with respect to the considered knowledge bases. After that, a new measure based on the exploitation of the taxonomical structure of a biomedical ontology is proposed. Using SNOMED CT as the input ontology, the accuracy of our proposal is evaluated and compared against other approaches according to a standard benchmark of manually ranked medical terms. The correlation between the results of the evaluated measures and the human experts’ ratings shows that our proposal outperforms most of the previous measures avoiding, at the same time, some of their limitations.

Introduction: The K4, Care European Project proposes a standard model for home-care (HC) assistance of elderly chronic patients (ECP), coordinating all professionals and institutions involved in HC for proper assistance. The model is supported by a distributed multi agent platform (MAP), a metaphor of the HC itself. Every person involved is represented by an intelligent agent interacting with other agents, representing other colleagues. Warnings are sent to the corresponding agent when required by other agents for some action to complete a medical service. Correct transferring of the HC knowledge to the MAP is a key issue. Methods: The actor profile ontology (APO) was build to determine the behaviour of MAP: the services to be provided have been specified by combinations of simpler actions; liabilities, responsibilities and access rights of every kind of professional involved in HC is specified, as well as interaction between different professionals. Agents are created according to APO definitions. Results: A prototype is implemented where physicians, nurses, social workers, carers... coordinate through their personal agents for HC of ECP, according to APO. The platform is currently being tested with real patients in Pollenza (It). Conclusion: Ontologies allow knowledge transmission to a computer application. APO provides high flexibility to mid and log-term evolution of HC, as changing it automatically changes the system behaviour. The basic structure of an ontology to support distributed medical complex systems is identified and this structure is easily exportable to other kind of medical systems, like hospital management or provision of dependency services.

Sequences of categorical data are in common use to represent sequences of events. In order to transfer such data to third parties for their analysis, masking methods can be applied to satisfy privacy laws and avoid the disclosure of sensitive information. Masking methods distort the data so that privacy is kept at the expenses of some information loss. %Different methods exist, each one trying to find a good trade-off between the risk of disclosure and the information loss. Microaggregation is one of the existing masking methods. In microaggregation small clusters are automatically built and the values of the members of a cluster are substituted by the values of the prototype of that cluster. Due to the fact that microaggregation is an NP-hard problem, heuristic approaches have been developed. Existing methods are mainly devoted to numerical and categorical data. The extension of these methods to sequences of categorical data requires the definition of special algorithms for clustering and prototyping.Artificial Intelligence offers techniques and tools that are appropriate for symbolic data. As in our context the sequences are defined in terms of categorical (symbolic) values, such AI techniques are of special relevance. In this paper, we will use them to propose a new method for generating the prototype of a small group of sequences of categorical values. These results can later be used in e.g. microaggregation.

Knowledge tailoring is a powerful tool to customize a system according to the particular view and interests of the user, in such a way that usability of the system becomes significantly improved. In this paper, the tailoring of an actor profile ontology is discussed in the context of a HomeCare integrated system that is being designed and implemented in the EU K4Care project. The paper analyzes different ways of implementing the automatic tailoring of an ontology. A tool called ATAPO, which has been designed and implemented to assist the user in this automatic tailoring is presented.

Computation of semantic similarity between concepts is a very common problem in many language related tasks and knowledge domains. In the biomedical field, several approaches have been developed to deal with this issue by exploiting the knowledge available in domain ontologies (SNOMED-CT) and specific, closed and reliable corpuses (clinical data). However, in recent years, the enormous growth of the Web has motivated researchers to start using it as the base corpus to assist semantic analysis of language. This paper proposes and evaluates the use of the Web as background corpus for measuring the similarity of biomedical concepts. Several classical similarity measures have been considered and tested, using a benchmark composed by biomedical terms and comparing the results against approaches in which specific clinical data were used. Results shows that the similarity values obtained from the Web are even more reliable than those obtained from specific clinical data, manifesting the suitability of the Web as an information corpus for the biomedical domain.

Healthcare applications involve complex structures of interacting processes and professionals that need to exchange information to provide the care services. In this kind of systems many different professional competencies, ethical and sensibility requirements as well a legal frameworks coexist and because of that the information managed inside the system should not be freely accessed, on the contrary it must be subject to very complex privacy restrictions. This is particularly critical in distributed systems, where additionally, security in remote transmissions must be ensured. In the paper we address the fundamental security issues that must be considered in design of a distributed healthcare application. The particular case of the K4CARE system is presented as an example of an application to the medical domain of homecare assistance. Homecare involves professionals from different institutions (hospital, social work organisms, etc) structurally independent, which must interact around any particular patient, and which use to be located in different physical places having their own and independent information systems.

One of the tasks towards the definition of a knowledge model for home care is the definition of the different roles of the users involved in the system. The roles determine the actions and services that can or must be performed by each type of user. In this paper the experience of building an ontology to represent the home-care users and their associated information is presented, in a proposal for a standard model of a Home-Care support system to the European Community.