Search

In the last few years, text analysis has grown as a keystone in several domains for solving many real-world problems, such as machine translation, spam detection, and question answering, to mention a few. Many of these tasks can be approached by means of machine learning algorithms. Most of these algorithms take as input a transformation of the text in the form of feature vectors containing an abstraction of the content. Most of recent vector representations focus on the semantic component of text, however, we consider that also taking into account the lexical and syntactic components the abstraction of content could be beneficial for learning tasks. In this work, we propose a content spectral-based text representation applicable to machine learning algorithms for text analysis. This representation integrates the spectra from the lexical, syntactic, and semantic components of text producing an abstract image, which can also be treated by both, text and image learning algorithms. These components came from feature vectors of text. For demonstrating the goodness of our proposal, this was tested on text classification and complexity reading score prediction tasks obtaining promising results. [ABSTRACT FROM AUTHOR]

The Semantic Web provides guidelines for the representation of information about real-world objects (entities) and their relations (properties). This is helpful for the dissemination and consumption of information by people and applications. However, the information is mainly contained within natural language sentences, which do not have a structure or linguistic descriptions ready to be directly processed by computers. Thus, the challenge is to identify and extract the elements of information that can be represented. Hence, this article presents a strategy to extract information from sentences and its representation with Semantic Web standards. Our strategy involves Information Extraction tasks and a hybrid semantic similarity measure to get entities and relations that are later associated with individuals and properties from a Knowledge Base to create RDF triples (Subject–Predicate–Object structures). The experiments demonstrate the feasibility of our method and that it outperforms the accuracy provided by a pattern-based method from the literature. [ABSTRACT FROM AUTHOR]

Creating effective mechanisms to detect misogyny online automatically represents significant scientific and technological challenges. The complexity of recognizing misogyny through computer models lies in the fact that it is a subtle type of violence, it is not always explicitly aggressive, and it can even hide behind seemingly flattering words, jokes, parodies, and other expressions. Currently, it is even difficult to have an exact figure for the rate of misogynistic comments online because, unlike other types of violence, such as physical violence, these events are not registered by any statistical systems. This research contributes to the development of models for the automatic detection of misogynistic texts in Latin American Spanish and contributes to the design of data augmentation methodologies since the amount of data required for deep learning models is considerable. [ABSTRACT FROM AUTHOR]

We provide a comprehensive survey of the research literature that applies Information Extraction techniques in a Semantic Web setting. Works in the intersection of these two areas can be seen from two overlapping perspectives: using Semantic Web resources (languages/ontologies/knowledge-bases/tools) to improve Information Extraction, and/or using Information Extraction to populate the Semantic Web. In more detail, we focus on the extraction and linking of three elements: entities, concepts and relations. Extraction involves identifying (textual) mentions referring to such elements in a given unstructured or semi-structured input source. Linking involves associating each such mention with an appropriate disambiguated identifier referring to the same element in a Semantic Web knowledge-base (or ontology), in some cases creating a new identifier where necessary. With respect to entities, works involving (Named) Entity Recognition, Entity Disambiguation, Entity Linking, etc. in the context of the Semantic Web are considered. With respect to concepts, works involving Terminology Extraction, Keyword Extraction, Topic Modeling, Topic Labeling, etc., in the context of the Semantic Web are considered. Finally, with respect to relations, works involving Relation Extraction in the context of the Semantic Web are considered. The focus of the majority of the survey is on works applied to unstructured sources (text in natural language); however, we also provide an overview of works that develop custom techniques adapted for semi-structured inputs, namely markup documents and web tables. [ABSTRACT FROM AUTHOR]

A common task in data analysis is to find the appropriate data sample whose properties allow us to infer the parameters and behavior of the data population. In data mining this task makes sense since usually the population is significantly huge, and thus it is required (for practical reasons) to obtain a subset that preserves its properties. In this regard, statistics offers some sampling techniques usually based on asymptotic results from the Central Limit Theorem. The effectiveness of such ways is bounded by several considerations as the sampling strategy (simple with or without replacement, stratified, cluster-based, etc.), the size of the population and the dimensionality of the space of the data. Due to these considerations alternative proposals are necessary. We propose a method based on a measure of information in terms of Shannon's Entropy. Our idea is to find the optimal sample whose information is as similar as possible to the information of the population, subject to several constraints. Finding such sample represents a hard optimization problem whose feasible space disallows the use of traditional optimization techniques. To solve it, we resort to a breed of Genetic Algorithm called Eclectic Genetic Algorithm. We test our method with synthetic datasets; the results show that our method is suitable. For completeness, we used several datasets from real problems; the results confirm the effectiveness of our proposal and allow us to visualize different applications. Finally, we establish a baseline based on selection instance methods as a point reference to measure the effectiveness of our method. [ABSTRACT FROM AUTHOR]

Integrated health services are characterized by a high degree of collaboration and communication among health professionals, as well as a merge of political, administrative, and technical actions, which can allow the sharing of information among healthcare team members (physicians, nurses, managers, and other stakeholders) related to patient care, and access to hospital infrastructure and technology, within a patient-centered approach. In this paper, we propose a technological solution based on software agents, which allows supporting the management of collaborative processes in environments of dynamic collaborative networks. We present a methodology for coordinating healthcare services through collaborative processes to enable organizations providing integrated care services and continuous process improvements. This methodology includes methods based on Model-Driven Development, which enable the generation of executable process models and the code of the process-oriented agents, derived from conceptual models of collaborative processes. The methodology and methods are implemented and automated by software agents that enable the generation of the technological solution, at run-time on the platform. Furthermore, we propose an agent-based platform that enables organizations to negotiate collaboration agreements in electronic format to establish collaborative networks, as well as a definition of the collaborative processes to be executed. In this way, the proposed agent-based platform allows collaborations among heterogeneous and autonomous healthcare organizations focusing on the process-oriented integration, enabling us to provide integrated healthcare services. [ABSTRACT FROM PUBLISHER]

Domain ontologies facilitate the organization, sharing, and reuse of domain knowledge. The construction of ontologies from text deals with the extraction of concepts and relations from text collection. A huge challenge is the learning of more expressive ontologies which includes relations such as disjointness or equivalence between classes. In our work, we exploit a text collection written in English with occurrence of named entities giving information about individuals in an specific domain knowledge for generating axioms. The co-ocurrence in texts of named entities is exploited to establish axiomatic relations between classes such as subClassOf, disjointWith, and equivalentClass; as well as between individuals as sameIndividualAs and differentFrom by using the instanceOf relation. For this purpose, a Named Entity Recognition tool was used for the identification of instanceOf relation, the linguistic context where the classes co-occur was extracted for the identification of axioms. The experiments were made on a text collection about the tourist domain. The results show that named entities provide good evidence for the identification of axiomatic relations. [ABSTRACT FROM PUBLISHER]

Two novel techniques based on term expansion for domain-specific Word Sense Disambiguation are presented in this paper. A list of terms correlated with the ambiguous word is obtained by means of correlation methods such as Pointwise Mutual Information and Chi-square. The list of correlated terms is used for expanding the context in which an ambiguous word occurs, thus leading to a better number of terms which may be further used for determining the correct sense of the target ambiguous word. The proposed method has been tested in domain-specific corpora (Finance and Sports) and in one balanced corpus, the British National Corpus (BNC). The obtained results for domain-specific corpora are competitive (Finance) and ever better (Sports and BNC) than those reported in previous works. [ABSTRACT FROM PUBLISHER]

We study the use of nonparametricmulticompare statistical tests on the performance of simulated annealing (SA), genetic algorithm (GA), particle swarm optimization (PSO), and differential evolution (DE), when used for electroencephalographic (EEG) source localization. Such task can be posed as an optimization problem for which the referred metaheuristic methods are well suited. Hence, we evaluate the localization's performance in terms of metaheuristics' operational parameters and for a fixed number of evaluations of the objective function. In this way, we are able to link the efficiency of themetaheuristics with a common measure of computational cost. Our results did not show significant differences in themetaheuristics' performance for the case of single source localization. In case of localizing two correlated sources, we found that PSO (ring and tree topologies) and DE performed the worst, then they should not be considered in large-scale EEG source localization problems. Overall, themulticompare tests allowed to demonstrate the little effect that the selection of a particular metaheuristic and the variations in their operational parameters have in this optimization problem. [ABSTRACT FROM AUTHOR]

The data produced by sensors of IoT devices are becoming keystones for organizations to conduct critical decision-making processes. However, delivering information to these processes in real-time represents two challenges for the organizations: the first one is achieving a constant dataflow from IoT to the cloud and the second one is enabling decision-making processes to retrieve data from dataflows in real-time. This paper presents a cloud-based Web of Things method for creating digital twins of IoT devices (named sentinels).The novelty of the proposed approach is that sentinels create an abstract window for decision-making processes to: (a) find data (e.g., properties, events, and data from sensors of IoT devices) or (b) invoke functions (e.g., actions and tasks) from physical devices (PD), as well as from virtual devices (VD). In this approach, the applications and services of decision-making processes deal with sentinels instead of managing complex details associated with the PDs, VDs, and cloud computing infrastructures. A prototype based on the proposed method was implemented to conduct a case study based on a blockchain system for verifying contract violation in sensors used in product transportation logistics. The evaluation showed the effectiveness of sentinels enabling organizations to attain data from IoT sensors and the dataflows used by decision-making processes to convert these data into useful information. [ABSTRACT FROM AUTHOR]

Indoor navigation systems incorporating augmented reality allow users to locate places within buildings and acquire more knowledge about their environment. However, although diverse works have been introduced with varied technologies, infrastructure, and functionalities, a standardization of the procedures for elaborating these systems has not been reached. Moreover, while systems usually handle contextual information of places in proprietary formats, a platform-independent model is desirable, which would encourage its access, updating, and management. This paper proposes a methodology for developing indoor navigation systems based on the integration of Augmented Reality and Semantic Web technologies to present navigation instructions and contextual information about the environment. It comprises four modules to define a spatial model, data management (supported by an ontology), positioning and navigation, and content visualization. A mobile application system was developed for testing the proposal in academic environments, modeling the structure, routes, and places of two buildings from independent institutions. The experiments cover distinct navigation tasks by participants in both scenarios, recording data such as navigation time, position tracking, system functionality, feedback (answering a survey), and a navigation comparison when the system is not used. The results demonstrate the system's feasibility, where the participants show a positive interest in its functionalities. [ABSTRACT FROM AUTHOR]

The most common machine-learning methods solve supervised and unsupervised problems based on datasets where the problem's features belong to a numerical space. However, many problems often include data where numerical and categorical data coexist, which represents a challenge to manage them. To transform categorical data into a numeric form, preprocessing tasks are compulsory. Methods such as one-hot and feature-hashing have been the most widely used encoding approaches at the expense of a significant increase in the dimensionality of the dataset. This effect introduces unexpected challenges to deal with the overabundance of variables and/or noisy data. In this regard, in this paper we propose a novel encoding approach that maps mixed-type data into an information space using Shannon's Theory to model the amount of information contained in the original data. We evaluated our proposal with ten mixed-type datasets from the UCI repository and two datasets representing real-world problems obtaining promising results. For demonstrating the performance of our proposal, this was applied for preparing these datasets for classification, regression, and clustering tasks. We demonstrate that our encoding proposal is remarkably superior to one-hot and feature-hashing encoding in terms of memory efficiency. Our proposal can preserve the information conveyed by the original data. [ABSTRACT FROM AUTHOR]

The automatic extraction of geospatial information is an important aspect of data mining. Computer systems capable of discovering geographic information from natural language involve a complex process called geoparsing, which includes two important tasks: geographic entity recognition and toponym resolution. The first task could be approached through a machine learning approach, in which case a model is trained to recognize a sequence of characters (words) corresponding to geographic entities. The second task consists of assigning such entities to their most likely coordinates. Frequently, the latter process involves solving referential ambiguities. In this paper, we propose an extensible geoparsing approach including geographic entity recognition based on a neural network model and disambiguation based on what we have called dynamic context disambiguation. Once place names are recognized in an input text, they are solved using a grammar, in which a set of rules specifies how ambiguities could be solved, in a similar way to that which a person would utilize, considering the context. As a result, we have an assignment of the most likely geographic properties of the recognized places. We propose an assessment measure based on a ranking of closeness relative to the predicted and actual locations of a place name. Regarding this measure, our method outperforms OpenStreetMap Nominatim. We include other assessment measures to assess the recognition ability of place names and the prediction of what we called geographic levels (administrative jurisdiction of places). [ABSTRACT FROM AUTHOR]

Entity extraction and linking (EEL) is an important task of the Semantic Web that allows to identify real-world objects from text and associate them with their respective resources from a Knowledge Base. Thus, one purpose of the EEL task is to extract knowledge from text. In recent years, several systems have been proposed for addressing such a task in several domains, languages, and knowledge bases. In this sense, some systems that combine the benefits of varied EEL systems have been proposed in a kind of ensemble system (like in Machine Learning) to provide better performance and extractions than using a single system. However, there are no clear indications for the selection, configuration, and result integration of EEL systems in an ensemble setting. This paper proposes a framework for the integration of EEL systems by providing recommendations for the selection of systems, the configuration of input parameters, the execution of systems, and the final integration of results through a filtering strategy that measures the occurrence of entities and detects the overlapping of entities. Based on the proposed framework, we implemented a system using existing EEL systems (through publicly available APIs). The experiments were performed through the GERBIL framework. Our results demonstrate an improvement of the micro/macro- precision and recall of the implemented system regarding the selected individual EEL systems over seven datasets. • Entity extraction and linking systems are integrated by a framework proposal. • The framework is based on an ensemble setting to integrate the output of systems. • The final output of EEL systems is filtered through an entity frequency strategy. • System parameters are tuned through experiments performed over diverse datasets. [ABSTRACT FROM AUTHOR]