10 documents found, page 1 of 1
Explaining Machine Learning: A Deeper Look into Admission Prediction
Consoli, Bernardo; Pedroso, Vinicius; Kniest, Artur; Vieira, Renata; Bordini, Rafael; Manssour, IsabelThe popularization of artificial intelligence solutions in both research and industry that has been occurring due to the rise of tools such as the GPT, Gemini and Claude large language models has revitalized research in the area. There are many possible uses within the medical field, but a key determinant of the adoption of new tools by medical professionals is trust. To augment tool trust, the tool must be mad...
Predicting Inpatient Admissions in Brazilian Hospitals
Consoli, Bernardo; Vieira, Renata; Bordini, Rafael; Manssour, IsabelPatient length-of-stay prediction is a topic of interest for hospital administrators, as it can aid in planning and the allocation of critical resources. Ideal resource allocation can result in better care and reduced costs. Artificial Intelligence solutions have been tested for this purpose using several datasets for both foreign and Brazilian hospitals, but focusing on long-term inpatient care or Intensive Ca...
Benchmarking the BRATECA Clinical Data Collection for Prediction Tasks
Consoli, Bernardo; Vieira, Renata; Bordin, RafaelExpanding the usability of location-specific clinical datasets is an important step toward expanding research into national medical issues, rather than only attempting to generalize hypotheses from foreign data. This means that benchmarking such datasets, thus proving their usefulness for certain kinds of research, is a worth- while task. This paper presents the first results of widely used prediction tasks fro...
Semantic Textual Similarity for Abridging Clinical Notes in Brazilian Electroni...
Bandeira, Lucas; Consoli, Bernardo; Vieira, Renata; Bordini, RafaelWith the growing importance of the use of information from electronic patient records in the development of machine learning models, there is also a need for a holistic understanding of those records, in particular abridging the clinical notes so that important information is used in the training process without the repetition that is commonly found in such notes. This paper presents the pre-processing of clini...
Enriching Portuguese Word Embeddings with Visual Information
Consoli, Bernardo; Vieira, RenataThis work focuses on the enrichment of existing Portuguese word embeddings with visual information. The combined text-image embeddings were tested against their text-only counterparts in common NLP tasks. The new embeddings were tested in two different domains - general news and a geosciences. The results show an increase in perfor- mance for several tasks, which indicates that visual information fusion for wor...
BRATECA (Brazilian Tertiary Care Dataset): a Clinical Information Dataset for t...
Consoli, Bernardo; Dias, Henrique; Vieira, Renata; Bordini, Rafael; Ana, UlbrichComputational medicine research requires clinical data for training and testing purposes, so the development of datasets composed of real hospital data is of utmost importance in this field. Most such data collections are in the English language, were collected in anglophone countries, and do not reflect other clinical realities, which increases the importance of national datasets for projects that hope to posi...
Enriching Portuguese Word Embeddings with Visual Information
Consoli, Bernardo; Vieira, RenataThis work focuses on the enrichment of existing Portuguese word embeddings with visual information in the form of visual embeddings. This information was extracted from images portraying given vocabulary terms and imagined visual embeddings learned for terms with no image data. These enriched embeddings were tested against their text-only counterparts in common NLP tasks. The results show an increase in perform...
Portuguese word embeddings for the oil and gas industry: Development and evalua...
Gomes, Diogo; Cordeiro, Fábio; Consoli, Bernardo; Santos, Nikolas; Moreira, Viviane; Vieira, Renata; Moraes, Silvia; Evsukoff, AlexandreOver the last decades, oil and gas companies have been facing a continuous increase of data collected in unstructured textual format. New disruptive technologies, such as natural language processing and machine learning, present an unprecedented opportunity to extract a wealth of valuable information within these documents. Word embedding models are one of the most fundamental units of natural language processi...
Embeddings for Named Entity Recognition in Geoscience Portuguese Literature
Consoli, Bernardo; Santos, Joaquim; Gomes, Diogo; Cordeiro, Fabio; Vieira, Renata; Moreira, VivianeThis work focuses on Portuguese Named Entity Recognition (NER) in the Geology domain. The only domain-specific dataset in the Portuguese language annotated for Named Entity Recognition is the GeoCorpus. Our approach relies on Bidirecional Long Short-Term Memory - Conditional Random Fields neural networks (BiLSTM-CRF) - a widely used type of network for this area of research - that use vector and tensor embeddin...
Word Embedding Evaluation in Downstream Tasks and Semantic Analogies
Santos, Joaquim; Consoli, Bernardo; Vieira, RenataLanguage Models have long been a prolific area of study in the field of Natural Language Processing (NLP). One of the newer kinds of language models, and some of the most used, are Word Embeddings (WE). WE are vector space representations of a vocabulary learned by a non-supervised neural network based on the context in which words appear. WE have been widely used in downstream tasks in many areas of study in N...