AI evidence extraction

ARTIFICIAL INTELLIGENCE AND OPTIMIZATION WITH THE METHOD REVIEW PROCESS INTEGRATED

David Lopes Maciel, Natiele Vieira de Oliveira Maciel, Aline Mikos, Brício Rocha Borges de Almeida, Paulo de Tarso Carvalho de Oliveira, Adjalma Campos de França Neto et al. · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

9/10

Relevance

2/4

Quality (LMQS)

Evidence

Citations

0.00

FWCI

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.22533/at.ed.394122608015

Methodology & findings

Study design

Case study demonstrating a five-phase systematic review methodology (PRIA: Integrated Review Process with Artificial Intelligence) applied to sustainability in native forest restoration projects in the Legal Amazon.

Main result

The practical application of the PRIA method demonstrated significant gains in efficiency and robustness. The study found that "During the search stage in the Web of Science (WoS) and Scopus databases, 1,408 articles were initially identified. After a rigorous process of eliminating duplicates and incomplete information, the final set comprised 1,382 valid articles." The analysis revealed that "interconnected articles often address the ecological, social, and economic dimensions in an integrated manner, highlighting emerging trends and pointing out important gaps for future investigations." Furthermore, "the potential of permanent nurseries as effective solutions for ecological restoration became evident, directly aligning with the Sustainable Development Goals (SDGs) and international commitments, especially the Paris Agreement."

Research paradigm

positivist/empiricist with pragmatic application focus

Author conclusions

"The PRIA method proved to be effective in conducting detailed and consistent systematic reviews. It stands out particularly for integrating open and accessible technologies, complemented by artificial intelligence, resulting in significant time savings while maintaining a high level of precision and scientific rigor." The authors further conclude that "the PRIA method constitutes an innovative and widely applicable approach, especially recommended for researchers interested in optimizing the efficiency, precision, and quality of their scientific outputs. Moreover, it promotes greater accessibility and inclusion in the academic landscape, democratizing the use of advanced technologies in various research contexts."

Risk of bias

Selection bias: Articles with invalid or incomplete data were automatically excluded, potentially removing relevant but poorly formatted studies; Database bias: Reliance on Web of Science and Scopus may exclude relevant articles from other databases or non-indexed sources; Language bias: Search limited to English and Brazilian Portuguese only; Publication bias: Preference for peer-reviewed articles may exclude relevant gray literature; Citation bias: Use of citation count and journal impact factor as screening criteria may bias toward well-established research; DOI requirement bias: Exclusion of 18 records without DOI, regardless of relevance; No explicit discussion of selection bias, attrition, or confounders; Reliance on database-specific coverage (WoS and Scopus may have language/publication type biases); Manual screening in Phase 4 after automated processes introduces subjective interpretation bias; Limited transparency on AI tool decision-making (Gemini, Litmaps algorithms not specified); No mention of inter-rater reliability or validation of tool outputs

Open questions raised

The authors identify that their practical application revealed 'important gaps for future investigations' in the literature on sustainability in native forest restoration projects in the Legal Amazon
The analysis highlighted the strategic relevance of permanent nurseries as effective solutions for ecological restoration, pointing to areas needing further investigation
The paper implicitly suggests that the PRIA method itself needs testing across diverse fields of knowledge to validate its generalizability
The study identifies gaps in the existing literature on sustainability in native forest restoration projects in the Legal Amazon. The authors note that their analysis "point[s] out important gaps for future investigations" in how ecological, social, and economic dimensions are integrated in restoration efforts.
Limited empirical validation of PRIA across diverse research fields beyond the Amazon restoration case study
Need for comparative studies with traditional systematic review methods to quantify efficiency gains

Data: Unified database containing 1,382 valid articles from Web of Science and Scopus (not explicitly made available); CSV file format mentioned: abstracts.bib (referenced as generated during Phase 3 preprocessing); Web of Science search results; Scopus search results; Unified database with abstracts; GitHub link to R script (https://drive.google.com/file/d/1Ko01nKgzv-_6HZlX89FliKaiuHg1_dPk/view?usp=sharing); CAPES journal portal access required for databases (https://www.periodicos.capes.gov.br/); Example Gemini queries with shareable links provided in Phase 1 activitiesCode: R script for data preprocessing available at: https://drive.google.com/file/d/1Ko01nKgzv-_6HZlX89FliKaiuHg1_dPk/view?usp=sharing; R script for Bibliometrix processing; Google Drive R script repository (https://drive.google.com/file/d/1Ko01nKgzv-_6HZlX89FliKaiuHg1_dPk/view?usp=sharing)Extracted from: pdfAgreement 56%

Explore related topics

Systematic Review Automation Literature Discovery Literature Review