• Publications
  • Influence
A systematic literature review on Natural Language Processing in Business Process Identification and Modeling
TLDR
The study presents techniques applied to the BPM life-cycle phases of Process Identification, Process Discovery and Process Analysis and tools to support Process Discovery to support research in extraction of business process models from natural language text.
Errors Analysis at the Zynq SoC ARM Embedded Processor Work presented in partial fulfillment of the requirements for the degree of Bachelor in Computer
TLDR
An analysis of the occurrence of software errors at the Xilinx Zynq-7000 programmable SoC, which adopts two ARM processing cores (more specifically a Cortex-A9 MPx2), which consists into comparing fault-injected executions with a fault-free application execution.
Study of a Deep Learning Approach to Named Entity Recognition for Portuguese
TLDR
This work proposes the addition of part-of-speech tagging to a system composed of a bidirectional LSTM and a CNN neural network architecture and evaluates the new system under the guidelines of the HAREM contest.
Evaluation of Arithmetic Operators for the MPPA-256 Manycore Architecture
TLDR
This work describes the internship at Kalray where some arithmetic operators’ architectures aiming for this type of processor were explored and analyzed, and the adopted methodology is described with the developed tools to support all analysis.
Fast Extract with Cube Hashing Work presented in partial fulfillment of the requirements for the degree of Bachelor in Computer
TLDR
A new implementation of the fast-extract algorithm, fxch, having complexity linear in the number of cubes is described, achieving a reduction in complexity by hashing sub-cubes and using the hash table to find good factors to extract.
Function Reuse on a Multi-Core VLIW Soft-Core Processor Work presented in partial fulfillment of the requirements for the degree of Bachelor in Computer Engineering
TLDR
This work proposes a low-cost hardware-based function reuse mechanism, which can optimize software execution by saving the inputs and outputs of the most recurring functions in a BRAM-based reuse table, so they can be reused in the next function calls, skipping actual execution, and improving performance.
COPA : A wireline / wireless convergent network monitoring and container manager architecture
TLDR
COPA, a wireless/wireline network convergent monitoring and container manager architecture for testbeds, is designed to support the research and development solutions for future computer networks and concludes that COPA provide a excellent environment for researches of wireless/ wireline convergent network and orchestration algorithms for containerized services.
Combining Performance and Diversity Measures for Optimizing Classification Ensembles via a Genetic Algorithm in the miRNA-Target Prediction Problem
TLDR
The purpose of this work is to study the effect of combining performance and diversity measures in a Genetic Algorithm’s (GA) fitness function that learns the best combination of classifiers in an heterogeneous ensemble classifier in the miRNA-Target prediction problem.
AnHALytics : a Workflow for Text Mining Trabalho
TLDR
Simulation results have shown that scientific workflows – created at first for applications from the biosciences domain – can also be used with text mining applications and present similar results, giving the application significant improvements compared to the classic scripting version of the workflow.