Leveraging AI and Data Visualization for Enhanced Policy-Making: Aligning Research Initiatives with Sustainable Development Goals DOI Open Access
Maicon Herverton Lino Ferreira da Silva, Leonides Medeiros Neto, Guto Leoni Santos

и другие.

Sustainability, Год журнала: 2024, Номер 16(24), С. 11050 - 11050

Опубликована: Дек. 17, 2024

Scientists, research institutions, funding agencies, and policy-makers have all emphasized the need to monitor prioritize investments outputs support achievement of United Nations Sustainable Development Goals (SDGs). Unfortunately, many current historic publications, proposals, grants were not categorized against SDGs at time submission. Manual post hoc classification is time-consuming prone human biases. Even when classified, few tools are available decision makers for supporting resource allocation. This paper aims develop a deep learning classifier categorizing abstracts by system policy-makers. First, we fine-tune Bidirectional Encoder Representations from Transformers (BERT) model using dataset 15,488 authors leading Brazilian universities, which preprocessed balanced training testing. Second, present PowerBI dashboard that visualizes classifications informed allocation sustainability-focused research. The achieved an F1-score, precision, recall exceeding 70% certain classes successfully classified existing projects, thereby enabling better tracking Agenda 2030 progress. Although capable classifying any text, it specifically optimized due nature its fine-tuning data.

Язык: Английский

Recent Advances in Polyvinylidene Fluoride with Multifunctional Properties in Nanogenerators DOI Open Access
Yueming Hu, Feijie Wang,

Yan Ma

и другие.

Small, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Abstract Amid the global energy crisis and rising emphasis on sustainability, efficient harvesting has become a research priority. Nanogenerators excel in converting abundant mechanical thermal into electricity, offering promising path for sustainable solutions. Among various nanogenerator's materials, Polyvinylidene fluoride (PVDF), with its distinctive molecular structure, exhibits multifunctional electrical properties including dielectric, piezoelectric pyroelectric characteristics. These combined excellent flexibility make PVDF prime candidate material nanogenerators. In nanogenerators, this is capable of efficiently collecting energy. This paper discusses how PVDF's are manifested three types nanogenerators compares performance these addition, strategies to improve output demonstrated, physical chemical modification as well structural optimization such hybrid structures external circuits. It also introduces application natural human harvesting, prospects medical technologies smart home systems. The aim promote use self‐powered sensing, monitoring, thereby providing valuable insights designing more versatile

Язык: Английский

Процитировано

0
On the road to urban sustainability: identifying major barriers to urban sustainability in Iran DOI
Hadi Alizadeh, Abolfazl Meshkini

Review of Regional Research, Год журнала: 2025, Номер unknown

Опубликована: Март 25, 2025

Язык: Английский

Процитировано

0
Leveraging AI and Data Visualization for Enhanced Policy-Making: Aligning Research Initiatives with Sustainable Development Goals DOI Open Access
Maicon Herverton Lino Ferreira da Silva, Leonides Medeiros Neto, Guto Leoni Santos

и другие.

Sustainability, Год журнала: 2024, Номер 16(24), С. 11050 - 11050

Опубликована: Дек. 17, 2024

Scientists, research institutions, funding agencies, and policy-makers have all emphasized the need to monitor prioritize investments outputs support achievement of United Nations Sustainable Development Goals (SDGs). Unfortunately, many current historic publications, proposals, grants were not categorized against SDGs at time submission. Manual post hoc classification is time-consuming prone human biases. Even when classified, few tools are available decision makers for supporting resource allocation. This paper aims develop a deep learning classifier categorizing abstracts by system policy-makers. First, we fine-tune Bidirectional Encoder Representations from Transformers (BERT) model using dataset 15,488 authors leading Brazilian universities, which preprocessed balanced training testing. Second, present PowerBI dashboard that visualizes classifications informed allocation sustainability-focused research. The achieved an F1-score, precision, recall exceeding 70% certain classes successfully classified existing projects, thereby enabling better tracking Agenda 2030 progress. Although capable classifying any text, it specifically optimized due nature its fine-tuning data.

Язык: Английский

Процитировано

0