Name: FLÁVIO IZO
Publication date: 06/05/2025
Examining board:
Name |
Role |
|---|---|
| CLAUDINE SANTOS BADUE | Examinador Interno |
| ELIAS SILVA DE OLIVEIRA | Presidente |
| FABIELLE CASTELAN MARQUES | Examinador Externo |
| HILÁRIO TOMAZ ALVES DE OLIVEIRA | Examinador Externo |
| HUGO RICARDO GONÇALO OLIVEIRA | Examinador Externo |
Summary: The constant growth in the publication of scientific documents poses enormous challenges
for the computing field in the activity of retrieving and extracting relevant information to
support research and innovation. To produce knowledge, we can consult several types of
documents, among which patents stand out. This type of document is an excellent source
of information and demonstrates the evolution of technical knowledge through the results
of its publications. It is estimated that approximately 70% of the information disclosed
in patent documents has never been published anywhere else. Thinking at an academic
level, the course program, through the teaching plans, is a pedagogical instrument that
describes the syllabus, objectives, programmatic contents, the evaluation process and the
primary bibliography of the curricular components present in the school curriculum. In
this way, the teaching plan serves as a reference for teachers and students to develop their
educational objectives, providing a basic curricular structure for the contents covered in
the course and that promote the teaching-learning process. This educational roadmap is
normally designed to promote a connection between the contents and the school activity
to be performed by the student. In this context, bringing patents closer to educational
institutions (EIs) has been highly encouraged. Collaboration between the two spheres
aims to ensure that students acquire the skills and knowledge to put into practice what
they learn in theory. On the other hand, it also allows industrial projects or services
to be improved within EIs. It is necessary to process and analyze the respective texts
to bridge patents and teaching plans. However, the difficulty in manually identifying
and analyzing the information in these documents makes it almost impossible to access
specific content, process the information and generate reports, especially when dealing
with large volumes of data and unstructured texts. Therefore, we propose an automated
methodology using Natural Language Processing (NLP) to bring the content of course
programs (teaching plans) closer to patent documents. The main objective is to assess
whether the courses offered cover the technical and scientific knowledge necessary for the
creation, development or improvement of patents. This study provides insights for EIs,
indicating content that can be included in their curricula to better prepare students for
the challenges of applied research and technological innovation while strengthening the
connection with the industrial sector. The results of this study, based on comparative
metrics, reveal that it is possible to verify whether teaching plans are aligned with the
demands of the industry and to inform where there are gaps that can be filled to improve
students’ ability to contribute to the development of patentable technological innovations.
