The rapid evolution and widespread accessibility of non-invasive medical imaging technologies, exemplified by Magnetic Resonance Imaging (MRI) and Computerized Tomography (CT), are fundamentally reshaping medical decisionmaking paradigms. These sophisticated imaging modalities, capable of extracting high-definition medical images, have emerged as integral components of modern healthcare, facilitating precise diagnostics and treatment planning. The escalating adoption of such technologies, however, has accentuated the need for a nuanced understanding and optimization of the performance and productivity of both medical equipment and the teams operating them, mainly due to the high costs and risks caused by their misuse. This work proposes using univariate analytical models to estimate the number of exams performed per day with machine learning algorithms. For such, different energy-related sensors monitoring 25 magnetic resonance equipment from three different brands were considered. The results of the research reveal a compelling validation of the proposed approach. A notably high Pearson correlation coefficient is observed between the predictions generated by the evaluated models and the real measurements obtained through the Radiology Information System (RIS). This robust correlation emphasizes the accuracy and reliability of the estimation models, validating their potential applicability in real-world healthcare scenarios. Furthermore, the study unveils an intriguing trend that distinguishes the performance of electric current sensors. Thirteen out of the 25 evaluated MRI machines demonstrate superior results when equipped with electric current sensors compared to other sensor types. This nuanced insight not only substantiates the critical role of energy-related sensors in predicting equipment performance but also underscores the importance of tailoring monitoring strategies to the unique characteristics of each machine.

Texts produced by the Brazilian judiciary have a complex and technical vocabulary, with elaborate use of the Portuguese language and many legal terms difficult to be understood, generating a barrier in communication between the judiciary and the population. In this sense, the Automatic Text Simplification (ATS), activity of the Natural Language Processing (NLP) area, can be applied to improve the readability of these types of text using specialized algorithms, and promote scalability in simplifying them, in view of the great demand in the courts. In this context, this article presents an evaluation of four methods of state of the art in text simplification, evaluated according to readability metrics, to improve the quality of existing texts in the judicial summaries, dataset containing 100 summaries of the Federal Regional Court of the 5th Region (TRF5) and another 100 of the Federal Supreme Court (STF). The methods MUSS(EN), MUSS(PT), Transformers and NMT + Attention were tested, and the results of the simplifications exceeded the FRE readability index of the original texts, making them more readable.

The sequencing of sampling algorithms has shown to be a promising approach in generating balanced versions of unbalanced data. Sequencing allows different algorithms of under-sampling and/or over-sampling to be performed in sequence, producing a resulting balanced database. However, defining the most appropriate sequence of sampling algorithms is challenging. This article treats the sequencing problem as a combinatorial optimization task and proposes a multi-objective optimization method to seek promising solutions that maximize the performance of classifiers both in accuracy and in F1-score. The results showed that the proposed method was capable of finding optimized sequences that improved the performance of the classifiers, obtaining statistically better results, mainly in F1- score, when compared with competing methods, in most of the selected unbalanced problems.

Domestic violence has increased globally as the COVID-19 pandemic combines with economic and social stresses. Some works have used traditional feature extractors to identify body positions to detect physical violence. Besides, the use of Machine Learning is limited by the trade-off between collecting more data while keeping users' privacy. Federated Learning (FL) is a technique that allows the creation of client-server networks, in which anonymized training data can be uploaded to a central model, responsible for aggregating and keeping the model up to date, and then distributing the updated model to the clients' nodes. This paper proposed an FL approach to the violence detection problem in video. The framework was evaluated on AIRTLab Dataset, in which frames were extracted. It used pretrained Convolutional Neural Networks (CNN) to address the image classification problem. Inception v3, MobileNet v2, ResNet-152 v2, and VGG-16 architectures were evaluated, with the MobileNet architecture presenting the best performance, in terms of accuracy (99.4%), with a loss of 0.5% when compared to the non-FL setting.

Roles that learners assume during online discussions are an important aspect of educational experience. The roles can be assigned to learners and/or can spontaneously emerge through student-student interaction. While existing research proposed several approaches for analytics of emerging roles, there is limited research in analytic methods that can i) automatically detect emerging roles that can be interpreted in terms of higher-order constructs of collaboration; ii) analyse the extent to which students complied to scripted roles and how emerging roles compare to scripted ones; and iii) track progression of roles in social knowledge progression over time. To address these gaps in the literature, this paper propose a network-analytic approach that combines techniques of cluster analysis and epistemic network analysis. The method was validated in an empirical study discovered emerging roles that were found meaningful in terms of social and cognitive dimensions of the well-known model of communities of inquiry. The study also revealed similarities and differences between emerging and script roles played by learners and identified different progression trajectories in social knowledge construction between emerging and scripted roles. The proposed analytic approach and the study results have implications that can inform teaching practice and development techniques for collaboration analytics.

Deep Convolutional Neural Networks (CNNs) have reached the attention in the last decade due to their successful application to many computer vision domains. Several handcrafted architectures have been proposed in the literature, with increasing depth and millions of parameters. However, the optimal architecture size and parameters setup are dataset-dependent and challenging to find. For addressing this problem, this work proposes a Multi-Objective Grammatical Evolution framework to automatically generate suitable CNN architectures (layers and parameters) for a given classification problem. For this, a Context-free Grammar is developed, representing the search space of possible CNN architectures. The proposed method seeks to find suitable network architectures considering two objectives: accuracy and F1-score. We evaluated our method on CIFAR-10, and the results obtained show that our method generates simpler CNN architectures and overcomes the results achieved by larger (more complex) state-of-the-art CNN approaches and other grammars.

Supervised machine learning models have been widely used to address the classification of messages in online discussions. Supervised learning algorithms require a large set of annotated data to accurately create a predictive model. However, data annotation is a complex task due to three factors: (i) depends on specialists to accurately label data; (ii) it is often a time-consuming and labour-intensive work,and(iii) in educational settings, it is not always easy to collect a substantial volume of data required by the machine learning algorithms. This paper presents an active learning-based approach that can reduce the amount of annotated data required to build machine learning models for the classification of educational data. The results obtained show that with only 20% of the annotated data, the proposed approach achieved similar results to those presented in the previous works that used the complete databases to train the machine learning model.

According to the World Health Organization, an estimated one billion people live with a disability. Millions of them are non-verbal and also experience motor-skill challenges. The restrictions on participation and communication caused by such disabilities often lead to discrimination and social exclusion, including the lack of access to formal education. Augmentative and Alternative Communication (AAC) is a method to afford communication for people with speech impairment. Software applications that implement AAC bring benefits of adaptability and personalization over traditional paper-based methods, but their usability needs improvement, particularly to increase user autonomy. This paper presents an interface redesign of the Livox AAC application, and a new user onboarding process based on user research to adjust the interface to user needs, contributing to user autonomy on AAC use.

The assessment of machine learning algorithms in a particular task is usually done by means of empirical evaluation on real world observational data. However, sometimes there is no previously annotated data available. Synthetic datasets have gained attention as an alternative for efficient classifier evaluation, since they are accessible and high parameterizable for a given learning task. The characterization of such databases can be done by means of descriptors on the learning object at hand, e.g., complexity measures, which extract statistical and geometric characteristics from the data sets, for a given classification problem, in order to estimate their complexity. Such complexity measures can be used to guide the production of synthetic datasets, so it reinforces one or more dataset features. The present work proposes the use of a many-objective algorithm for the generation of synthetic data considering four measures of complexity that will be balanced at the same time. The results showed that the proposal is able to optimize conflicting objectives, generating datasets of specific complexities.

Feedback is a crucial element in helping students identify gaps and assess their learning progress. In online courses, feedback becomes even more critical as it is one of the resources where the teacher interacts directly with the student. However, with the growing number of students enrolled in online learning, it becomes a challenge for instructors to provide good quality feedback that helps the student self-regulate. In this context, this paper proposed a content analysis of feedback text provided by instructors based on different indicators of good feedback. A random forest classifier was trained and evaluated at different feedback levels. The results achieved outcomes up to 87% and 0.39 of accuracy and Cohen's κ, respectively. The paper also provides insights into the most influential textual features of feedback that predict feedback quality.

This paper presents an approach to automate the selection and execution of previously indentified test cases for loading procedures in Business Intelligence (BI) environments based on Data Warehouse (DW).To verify and validate the approach, a unit test framework was developed.The overall goal is achieve data quality improvement.The specific aim is reduce test effort and, consequently, promote test activities in data warehousing process.A controlled experiment evaluation was carried out to investigate the adequacy of the proposed method for data warehouse procedures development.The results of the experiment show that our approach clearly reduces test effort when compared with manual execution of test cases.

BACKGROUND: Drug-target networks are receiving a lot of attention in late years, given its relevance for pharmaceutical innovation and drug lead discovery. Different in silico approaches have been proposed for the identification of new drug-target interactions, many of which are based on kernel methods. Despite technical advances in the latest years, these methods are not able to cope with large drug-target interaction spaces and to integrate multiple sources of biological information. RESULTS: We propose KronRLS-MKL, which models the drug-target interaction problem as a link prediction task on bipartite networks. This method allows the integration of multiple heterogeneous information sources for the identification of new interactions, and can also work with networks of arbitrary size. Moreover, it automatically selects the more relevant kernels by returning weights indicating their importance in the drug-target prediction at hand. Empirical analysis on four data sets using twenty distinct kernels indicates that our method has higher or comparable predictive performance than 18 competing methods in all prediction tasks. Moreover, the predicted weights reflect the predictive quality of each kernel on exhaustive pairwise experiments, which indicates the success of the method to automatically reveal relevant biological sources. CONCLUSIONS: Our analysis show that the proposed data integration strategy is able to improve the quality of the predicted interactions, and can speed up the identification of new drug-target interactions as well as identify relevant information for the task. AVAILABILITY: The source code and data sets are available at www.cin.ufpe.br/~acan/kronrlsmkl/.

Information Extraction (IE) aims to extract from textual documents only the fragments which correspond to datafields required by the user. In this paper, we present new experiments evaluating a hybrid machine learning approach for IE that combines text classifiers and Hidden MarkovModels (HMM). In this approach, a text classifier technique generates an initial output, which is refined by an HMM, taking into account dependences in the order of the data to be extracted. The proposal was evaluated to extract information from bibliographic references. Experiments performed on a corpus of 6000 references have shown an improvement in performance compared to benchmarking IE approaches adopted in previous work.