Collections are objects used to arrange, into a single unit, multiple data items that form a natural group. Different types of collections exist, due to different constraints based on whether or not they impose an order on their elements and whether or not they allow repetition of elements. Any of them are easily found in several domains of our everyday life. For instance, a deck of cards, the prime divisors of a number or the teams that compete in a championship can be seen as a collection. Thus, an effective modeling of collections is a recurring issue in information management. In the ontology design field, recurring modeling problems can be addressed by the use of Ontology Design Patterns (ODPs). In the case of collections, ODPs have been proposed for representing sequences, lists, sets and bags. However, none of these patterns are completely adequate for representing collections of ordered elements without repetition. In this paper we present an ODP for representing that notion, which we have named Permutation . Moreover, another ODP named ListOfPermutations is also introduced, which allows to represent how the order of a Permutation varies along time. Because not all constraints required by these ODPs can be represented in OWL 2, SHACL shapes have been used in their definitions. • ODP Permutation for representing ordered collections without repetition. • ODP ListOfPermutations for representing ordered permutations of the same elements. • Use of SHACL to define constraints of the proposed ODPs not representable in OWL 2.

The increasing interest among manufacturers in monitoring and analyzing industrial systems is generating a problem related to the considerable costs associated with the storage of captured data. This paper presents a three-level hierarchical architecture for time-series data storage on cloud environments that helps to decrease those costs. In the first level, new raw time-series data is stored for a short-period of time (e.g., one day) on electronic non-volatile storage such as solid-state drives (SSDs) that provide fast access for real time visualization of the latest data. In the second level, recent time series are stored for a medium-period of time (e.g., one week) on magnetic hard disk drives (HDDs) that are lower-cost devices with slower data transfer speed. In the third level, a reduced representation of the time series obtained by applying time-series reduction techniques are stored in HDDs, for a longer period of time (e.g., one year). Dealing with those reduced representations, data storage and transmission costs can be decreased, without limiting the future use of the data in different processes.The architecture has been implemented by using the top Database Management System from three different categories: Wide column store, Time series DBMS and Graph DBMS. It has been tested using industrial time series coming from a real manufacturing environment, and with three different types of queries proposed by domain experts. The performance results regarding storage space, storage costs and query time processing are shown on the paper.

Data pre-processing in smart manufacturing scenarios implies that data engineers must have knowledge of many different techniques necessary for example to clean, segment, normalize and to efficiently store time-series data. However, for many data engineers often results difficult to be aware of the most adequate pre-processing techniques due to the heterogeneous time-series data they must deal with and the variety of existing techniques.In this paper, a visual-interactive web system that facilitates the pre-processing task to data engineers is presented. On the one hand, data engineers can use it to select an adequate technology for detecting and handling outliers, to remove noise and to impute missing values in time series. Moreover, the system provides some recommendations to select the appropriate techniques and parameter values required by them. On the other hand, the system includes a machine learning module that suggests the most appropriate reduced representations in order to efficiently store those time series. The system has been tested with real time series of two different manufacturing companies.

The massive captured data from industrial sensors (time-series data) that could serve as relevant indicators for predictive maintenance of equipment, fault diagnosis, etc. is generating a problem related to the considerable costs associated with their storage.

BACKGROUND: has also been used to build these kinds of systems, but contradictory results have been reported: some works conclude that Kinect does not provide an accurate method of measuring gait kinematics variables, but others, on the contrary, report good accuracy results. METHODS: In this work, we have built a Kinect-based system that can distinguish between different PD stages, and have performed a clinical study with 30 patients suffering from PD belonging to three groups: early PD patients without axial impairment, more evolved PD patients with higher gait impairment but without Freezing of Gait (FoG), and patients with advanced PD and FoG. Those patients were recorded by two Kinect devices when they were walking in a hospital corridor. The datasets obtained from the Kinect were preprocessed, 115 features identified, some methods were applied to select the relevant features (correlation based feature selection, information gain, and consistency subset evaluation), and different classification methods (decision trees, Bayesian networks, neural networks and K-nearest neighbours classifiers) were evaluated with the goal of finding the most accurate method for PD stage classification. RESULTS: The classifier that provided the best results is a particular case of a Bayesian Network classifier (similar to a Naïve Bayesian classifier) built from a set of 7 relevant features selected by the correlation-based on feature selection method. The accuracy obtained for that classifier using 10-fold cross validation is 93.40%. The relevant features are related to left shin angles, left humerus angles, frontal and lateral bents, left forearm angles and the number of steps during spin. CONCLUSIONS: In this paper, it is shown that using Kinect is adequate to build a inexpensive and comfortable system that classifies PD into three different stages related to FoG. Compared to the results of previous works, the obtained accuracy (93.40%) can be considered high. The relevant features for the classifier are: a) movement and position of the left arm, b) trunk position for slightly displaced walking sequences, and c) left shin angle, for straight walking sequences. However, we have obtained a better accuracy (96.23%) for a classifier that only uses features extracted from slightly displaced walking steps and spin walking steps. Finally, the obtained set of relevant features may lead to new rehabilitation therapies for PD patients with gait problems.

A growing number of Linked Open Data sources (from diverse provenances and about different domains) that can be freely browsed and searched to find and extract useful information have been made available. However, access to them is difficult for different reasons. This study addresses access issues concerning heterogeneity. It is common for datasets to describe the same or overlapping domains while using different vocabularies. Our study presents a transducer that transforms a SPARQL query suitably expressed in terms of the vocabularies used in a source dataset into another SPARQL query suitably expressed for a target dataset involving different vocabularies. The transformation is based on existing alignments between terms in different datasets. Whenever the transducer is unable to produce a semantically equivalent query because of the scarcity of term alignments, the transducer produces a semantic approximation of the query to avoid returning the empty answer to the user. Transformation across datasets is achieved through the management of a wide range of transformation rules. The feasibility of our proposal has been validated with a prototype implementation that processes queries that appear in well-known benchmarks and SPARQL endpoint logs. Results of the experiments show that the system is quite effective in achieving adequate transformations.

Big Data is one of the key enabling technologies in smart manufacturing, where manufacturing companies aim at leveraging the data generated throughout their processes. The potential of Big Data Analytics is particularly significant in the context of manufacturing companies distributed worldwide. These companies own several manufacturing plants operating the same process in different environments and conditions. This generates massive amounts of data that could be analyzed in order to improve process efficiency and product quality. This paper presents the requirements for an architecture to capture, integrate and analyze the large-scale volumes of data generated in a real-world manufacturing business scenario - a chemical manufacturing sector distributed worldwide-. This scenario serves as a case study for an applied research project on Big Data Analytics. The business nature of this scenario provides those real-life requirements the architecture has to deal with. Existing approaches can be extended to fulfill these requirements, in order to be effectively applied in similar manufacturing business contexts.

One of the most promising areas where Big Data Analytics can be integrated into business-oriented projects-allowing research and development teams to work hand in hand with industry representatives - is the digitalization of manufacturing industry. There are two main driving forces for the interest in this area: the promotion of key strategies such as German Government's Industrie 4.0 or General Electric's Industrial Internet, and the use of servitization strategies to transform manufacturing business models. This paper presents a case study based on a Big Data Analytics applied research project developed for a capital equipment manufacturer. Their current business model is based on selling machinery and storage infrastructure for larger chemical manufacturing companies spread worldwide. The project is developed in the context of a servitization strategy where this capital equipment manufacturer aims at attaching valued-added services to their products, leveraging the use of Big Data Analytics to assist their customers in order to optimize their production process. The paper uses the CRISP-DM (CRoss-Industry Standard Process for Data Mining) methodology as a framework to organize and present our main findings related to the Business Understanding phase. This allow us to provide pragmatic, business-oriented considerations that can be leveraged by research and development teams when exploring opportunities to develop Big Data Analytics projects in the context of manufacturing servitization. Thus, they can face the initial steps of those projects with a better understanding of the specificities of this application field, as well as an a priori identification of problematic situations that may arise and required competencies to be covered by their team members.

The goal of this paper is to show the main features of KiReS, a telerehabilitation system based on Kinect for Windows, that offers, for both, users and physiotherapists some specific elements that make it more friendly to them. From the point of view of users, they can see in two 3D avatars how an exercise must be executed and how they execute it respectively. This feature can help them improve exercises performance. Moreover during the rehabilitation session they will always see an informative list that shows the exercises to be done in the session. From the point of view of physiotherapists the system allows them on the one hand, to define customized rehabilitation therapies. That can be done by defining different exercises that combine pre-defined movements. Moreover, they can add tests oriented to specific illnesses so that users themselves evaluate their physical state. On the other hand, they can create new exercises just performing those exercises in front of the system and recording them. Those features, not fully supported by already existing telerehabilitation systems, provide an added value that is well valued by both groups. Moreover, a prototype of KiReS is in operation, and allowed us to test its suitability from the point of view of real time performance as well as from the point of view of usability.

OBJECTIVES: To study ontology modularization techniques when applied to SNOMED CT in a scenario in which no previous corpus of information exists and to examine if frequency-based filtering using MEDLINE can reduce subset size without discarding relevant concepts. MATERIALS AND METHODS: Subsets were first extracted using four graph-traversal heuristics and one logic-based technique, and were subsequently filtered with frequency information from MEDLINE. Twenty manually coded discharge summaries from cardiology patients were used as signatures and test sets. The coverage, size, and precision of extracted subsets were measured. RESULTS: Graph-traversal heuristics provided high coverage (71-96% of terms in the test sets of discharge summaries) at the expense of subset size (17-51% of the size of SNOMED CT). Pre-computed subsets and logic-based techniques extracted small subsets (1%), but coverage was limited (24-55%). Filtering reduced the size of large subsets to 10% while still providing 80% coverage. DISCUSSION: Extracting subsets to annotate discharge summaries is challenging when no previous corpus exists. Ontology modularization provides valuable techniques, but the resulting modules grow as signatures spread across subhierarchies, yielding a very low precision. CONCLUSION: Graph-traversal strategies and frequency data from an authoritative source can prune large biomedical ontologies and produce useful subsets that still exhibit acceptable coverage. However, a clinical corpus closer to the specific use case is preferred when available.

Although the goal of achieving semantic interoperability of electronic health records (EHRs) is pursued by many researchers, it has not been accomplished yet. In this paper, we present a proposal that smoothes out the way toward the achievement of that goal. In particular, our study focuses on medical diagnoses statements. In summary, the main contributions of our ontology-based proposal are the following: first, it includes a canonical ontology whose EHR-related terms focus on semantic aspects. As a result, their descriptions are independent of languages and technology aspects used in different organizations to represent EHRs. Moreover, those terms are related to their corresponding codes in well-known medical terminologies. Second, it deals with modules that allow obtaining rich ontological representations of EHR information managed by proprietary models of health information systems. The features of one specific module are shown as reference. Third, it considers the necessary mapping axioms between ontological terms enhanced with so-called path mappings. This feature smoothes out structural differences between heterogeneous EHR representations, allowing proper alignment of information.

Broadcasting sport events in live is a challenging task because obtaining the best views requires taking into account many dynamic factors, such as: the location and movement of interesting objects, all the views provided by cameras in the scenario (some of them wireless, mobile, or attached to moving objects), possible occlusions, etc. Therefore, a technical director needs to manage a great amount of continuously changing information to quickly select the camera whose view should be broadcasted.

Nowadays it is widely recognized the need of developing a new generation of professional equipment for broadcasting that allows a considerable reduction of productions costs in different scenarios. We consider mobile multimedia scenarios where the incorporation of last generation hardware and the development of new software should allow: 1) to decrease the needs of human resources to produce (create, edit, and distribute) audiovisual content; and 2) to increase the richness of the broadcasted product.

Patients suspected of suffering sleep apnea and hypopnea syndrome (SAHS) have to undergo sleep studies such as expensive polysomnographies to be diagnosed. Healthcare professionals are constantly looking for ways to improve the ease of diagnosis and comfort for this kind of patients as well as reducing both the number of sleep studies they need to undergo and the waiting times. Relating to this scenario, some research proposals and commercial products are appearing, but all of them record the physiological data of patients to portable devices and, in the morning, these data are loaded into hospital computers where physicians analyze them by making use of specialized software. In this paper, we present an alternative proposal that promotes not only a transmission of physiological data but also a real-time analysis of these data locally at a mobile device. For that, we have built a classifier that provides an accuracy of 93% and a receiver operating characteristic-area under the curve (ROC-AUC) of 98.5% on SpO(2) signals available in the annotated Apnea-ECG Database. This local analysis allows the detection of anomalous situations as soon as they are generated. The classifier has been implemented taking into consideration the restricted resources of mobile devices.

Networks of sensors are used in many different fields, from industrial applications to surveillance applications. A common feature of these applications is the necessity of a monitoring infrastructure that analyzes a large number of data streams and outputs values that satisfy certain constraints.

Continuous pervasive monitoring has the potential to improve the quality of life of many persons (for example those who suffer from heart arrhythmias or elderly people who suffer from chronic diseases). We show the three main steps related to the monitoring process and supported by our system. First, how data are captured by sensors which send those data using wireless communications to a PDA (personal digital assistant). Second, how data are analyzed, locally on the PDA, using techniques proposed in the areas of the semantic Web and machine learning. Finally, how data stored at the PDA can be queried remotely using Web services. Those three steps are illustrated in two different scenarios that the system can deal with: tele-assistance and monitoring of arrhythmias. Moreover, through the paper we highlight the main advantages provided by the system: active monitoring which consists in reacting to anomalous situations without direct intervention of the user; universal assistance, i.e. irrespective of time or place through the use of wireless communications and PDAs; vital signs monitoring which consists in using sensors that capture the value of those vital signs, and subsequently feed them to a decision support system that analyses them and generates an alarm if necessary; and remote monitoring which allows authorized personal to consult data on monitored patients, using the Internet. Finally we present some performance results which demonstrate the feasibility of the system

In this paper we investigate if techniques based on decision trees are also useful to classify ischemia. In order to do that, we have based on a previous own algorithm that is able of detecting ST-segment episodes, can be executed in real time and is lightweight enough to be implemented on mobile devices such as PDAs. Three main steps are performed by that algorithm: 1) signal preprocessing that extracts important features from the ECG signal 2) detection of suspect ST segment events and 3) rejection of irrelevant ST segment events. We have found that techniques based on decision trees are interesting for the last step in order to obtain a set of rules that reject some of the suspect ST segment events detected in the second step. The freely available part of the LTST database has been used to develop the detector and the rest of the same database has been used for validation purposes. The sensitivity of the detector over all the records of the LTST database is 89.89% for episodes annotated according to C protocol and the positive predictivity is 70.03% for episodes annotated according to A protocol. Those results improve our previously obtained ones and we think that they are comparable to other results that appear in the literature. However, it has to be noticed that we have not found works that detect ischemia in real time and show validation results over the LTST database

The ability to retrieve software in an easy and efficient way confers competitive advantage on computer users in general and, even more especially, on users of wireless devices (like some laptops, PDAs, etc.). In this article, we present a software retrieval service that allows users to select and retrieve software in an easy and efficient way, anywhere and anytime. Two relevant components of this service are: 1) a software ontology (software catalog) which provides users with a semantic description of software elements, hiding the location and access method of various software repositories, and 2) a set of specialist agents that allow browsing of the software catalog (automatically customized for each user), and an efficient retrieval method for the selected software. These agents automatically adapt their behavior to different users and situations by considering the profile and preferences of the users and the network status.In summary, our software-obtaining process based on an ontology and autonomous and adaptive agents presents a qualitative advance with respect to existing solutions: our approach adapts to the features of users, relieving them from knowing the technical features of their devices and the location and access method of various remote software repositories.

With the current advances of mobile computing technology, we are witnessing an explosion in the development of applications that provide mobile users with a wide range of services. In this paper, we present a system that supports distributed processing of continuous location-dependent queries in mobile environments. The system that we propose presents the following main advantages: 1) it is a general solution for the processing of location-dependent queries in scenarios where not only the users issuing queries, but also other interesting objects can move; 2) it performs an efficient processing of these queries in a continuous way; 3) it is especially well adapted to environments where location data are distributed in a network and processing tasks can be performed in parallel, allowing a high scalability; and 4) it optimizes wireless communications. We use mobile agents to carry the processing tasks wherever they are needed. Thus, agents are in charge of tracking the location of interesting moving objects and refreshing the answer to a query efficiently. We evaluate the usefulness of the presented proposal showing that the system achieves a good precision and scales up well

The new advances in sensor technology, personal digital assistants (PDAs), and wireless communications favor the development of a new type of monitoring system that can provide patients with assistance anywhere and at any time. Of particular interest are the monitoring systems designed for people that suffer from heart arrhythmias, due to the increasing number of people with cardiovascular diseases. PDAs can play a very important role in these kinds of systems because they are portable devices that can execute more and more complex tasks. The main questions answered in this paper are whether PDAs can perform a complete electrocardiogram beat and rhythm classifier, if the classifier has a good accuracy, and if they can do it in real time. In order to answer these questions, in this paper, we show the steps that we have followed to build the algorithm that classifies beats and rhythms, and the obtained results, which show a competitive accuracy. Moreover, we also show the feasibility of incorporating the built algorithm into the PDA.

Among mobile computing applications, location-based services are expected to be a big business for wireless operators. These services provide value added by considering the location of the user in order to give him/her more customized information. However, the processing of continuous location-dependent queries is still a subject of research. One of the key issues is how to update the answer presented to the user efficiently as it becomes obsolete very quickly. The goal of this paper is to analyze how to deal with situations where the answer cannot be updated with the desired frequency. We present several approaches to this problem and evaluate their performance.

The World Wide Web is fast becoming a ubiquitous computing environment. Prevalent keyword-based search techniques are scalable, but are incapable of accessing information based on concepts. We investigate the use of concepts from multiple, real-world pre-existing, domain ontologies to describe the underlying data content and support information access at a higher level of abstraction. It is not practical to have a single domain ontology to describe the vast amounts of data on the Web. In fact, we expect multiple ontologies to be used as different world views and present an approach to "browse" ontologies as a paradigm for information access. A critical challenge in this approach is the vocabulary heterogeneity problem. Queries are rewritten using interontology relationships to obtain translations across ontologies. However, some translations may not be semantics preserving, leading to uncertainty or loss in the information retrieved. We present a novel approach for estimating loss of information based on the navigation of ontological terms. We define measures for loss of information based on intensional information as well as on well established metrics like precision and recall based on extensional information. These measures are used to select results having the desired quality of information.

The huge number of autonomous and heterogeneous data repositories accessible on the "global information infrastructure" makes it impossible for users to be aware of the locations structure/organization, query languages and semantics of the data in various repositories. There is a critical need to complement current browsing, navigational and information retrieval techniques with a strategy that focuses on information content and semantics. In any strategy that focuses on information content, the most critical problem is that of different vocabularies used to describe similar information across domains. We discuss a scalable approach for vocabulary sharing. The objects in the repositories are represented as intensional descriptions by preexisting ontologies expressed in Description Logics characterizing information in different domains. User queries are rewritten by using interontology relationships to obtain semantics preserving translations across the ontologies.

Due to the emerging interest in integrating different application environments, there have been many recent proposals for federated systems. In this paper, a federated system that permits the integration of heterogeneous relational databases using a terminological knowledge representation system is presented. In particular, two of the system's components: the translator and the integrator are explained in depth. The translator permits one to obtain a terminology from a relational schema, either semiautomatically, by expressing database properties, or manually, by using a set of predefined operations. In turn, the integrator generates a federated terminology by integrating several terminologies using the semantics expressed as correspondences between the data elements of different terminologies. Unlike many other approaches, the use of a terminological system permits us to obtain a semantically richer federated terminology and, at the same time, define a wider and more consistent integration process.

The processing cost of queries with an empty answer, both in the database and knowledge base context, is usually high. One purpose of semantic query optimization methods in the database context is to use semantic knowledge to detect such types of queries. Although semantic query optimization is well known in the database context, this is not the case for knowledge base systems (KBSs). This paper presents a method that allows the detection of queries with an empty answer using only semantic information expressed in the knowledge base definition. The method can be applied in the context of KBSs that provide some of the following features: structuring mechanisms, assertional knowledge, temporal information, and handling of inequality expressions.>

The authors describe a framework for query optimization for knowledge base management systems (KBMSs) based on the knowledge representation language Telos. The framework involves temporal and syntactic simplifications and semantic modification of the queries. Temporal simplification attempts to select parts of a knowledge base that are relevant to a query from a temporal viewpoint. Syntactic simplification exploits structural properties of the data model and attempts to transform the query into an equivalent and more efficient one. Semantic transformation uses knowledge specific to the application domain to transform a user-specified query into another form which gets the same answer and is processed efficiently. The three steps were integrated into a global algorithm for query optimization that utilizes all features of the considered KBMS.>