Scholarlink

At the Aeronautics Institute of Technology (ITA) in Brazil, two academic case studies focused on interdisciplinary problem-based learning (IPBL) were carried out. The first one named Fluency Analysis in Literacy Version 3.0 on Real Time (FAL3-RT) was developed on the 2nd Semester of 2023 and involved 13 undergrad and graduate students to address a problem of real-time system. The second one named Fluency Assessment in Literacy on Big Data (FAL-BD) was developed on the 1st Semestre of 2024 and involved 11 undergrad and graduate students to address a problem of Big Data and Large Language Models (LLM). These two case studies involved 24 undergrad and graduate students from seven different courses of the Electronic and Computer Engineering Graduate Program. Four from the 2nd Semester of 2023 and three from the 1st Semester of 2024. Both case studies aimed to conceptualize, model, develop, and test components of a distributed database system. Over the course of two semesters, a prototype for audio collection was created, leveraging technologies such as Database Systems, Artificial Intelligence, Real-Time Embedded Systems, Machine Learning, Blockchain, and Kubernetes. The final prototypes were based on an initial research project developed for the Brazilian Ministry of Education, focused on the automatic analysis of reading fluency in elementary school children. This paper collectively describes the computational infrastructure for real-time audio collection, implemented in the FAL3-TR embedded system, and its improvements in the FAL-BD using large language models (LLMs). The audio, collected via a developed interface, was stored on a server for subsequent automatic analysis using a machine learning model. Both projects were completed within 32 weeks of an academic year, employing the Scrum Framework for project management. The key contribution of this work was the integration of the Scrum Agile Methodology with various technologies to manage, develop, and test both case studies, culminating in the development of a Literacy Fluency Assessment system, including a functional software and hardware prototype for audio acquisition.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Case Study Involving Interdisciplinary Academic Projects for Reading Fluency Assessment and Analysis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Victor Araujo Paula Cavichioli,
  • Stephanie Souza Russo,
  • Huascar Mirko Montecinos Cortez,
  • Ana Carolina de Lima Angelo,
  • Gildarcio Sousa Goncalves,
  • Adilson Marques da Cunha,
  • Luiz Alberto Vieira Dias,
  • Lineu Fernando Stege Mialaret,
  • Johnny Cardoso Marques

摘要

At the Aeronautics Institute of Technology (ITA) in Brazil, two academic case studies focused on interdisciplinary problem-based learning (IPBL) were carried out. The first one named Fluency Analysis in Literacy Version 3.0 on Real Time (FAL3-RT) was developed on the 2nd Semester of 2023 and involved 13 undergrad and graduate students to address a problem of real-time system. The second one named Fluency Assessment in Literacy on Big Data (FAL-BD) was developed on the 1st Semestre of 2024 and involved 11 undergrad and graduate students to address a problem of Big Data and Large Language Models (LLM). These two case studies involved 24 undergrad and graduate students from seven different courses of the Electronic and Computer Engineering Graduate Program. Four from the 2nd Semester of 2023 and three from the 1st Semester of 2024. Both case studies aimed to conceptualize, model, develop, and test components of a distributed database system. Over the course of two semesters, a prototype for audio collection was created, leveraging technologies such as Database Systems, Artificial Intelligence, Real-Time Embedded Systems, Machine Learning, Blockchain, and Kubernetes. The final prototypes were based on an initial research project developed for the Brazilian Ministry of Education, focused on the automatic analysis of reading fluency in elementary school children. This paper collectively describes the computational infrastructure for real-time audio collection, implemented in the FAL3-TR embedded system, and its improvements in the FAL-BD using large language models (LLMs). The audio, collected via a developed interface, was stored on a server for subsequent automatic analysis using a machine learning model. Both projects were completed within 32 weeks of an academic year, employing the Scrum Framework for project management. The key contribution of this work was the integration of the Scrum Agile Methodology with various technologies to manage, develop, and test both case studies, culminating in the development of a Literacy Fluency Assessment system, including a functional software and hardware prototype for audio acquisition.