Study goals
Evaluate how the academic production that suggests the use of AI and Deep Learning to support sustainable project management considers data quality as an impact factor on the results of AI training
Relevance / originality
With the use of AI to support the activity of organizations, the quality of the data used in the training of algorithms acquires fundamental relevance, being necessary to understand if the proponents of this use have taken this aspect into account
Methodology / approach
Bibliometric study using Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Prisma to identify which aspects among five categories of data quality problems cited in the literature on AI and Deep Learning have been considered in proposals for its use
Main results
Data quality in its aspects of precision and accuracy is already present in all articles. New problems arising from the operation and training of AI are not yet considered, which may negatively impact this use and its results
Theoretical / methodological contributions
We propose the need for ongoing discussions on data quality in the academic community responsible for the development of AI algorithms to be accompanied by those responsible for the decision of their adoption in the management of sustainable projects
Social / management contributions
By deepening the analysis of the quality and representativeness of the data used in AI training, incorporating dimensions of socio-environmental sustainability, it becomes possible to achieve superior results in all its dimensions and not just in the economic aspect of sustainability