A government agency is operationalizing a new AI tool for predictive policing. The project manager needs to identify data subject matter experts (SMEs) to ensure data quality and relevance. The project team has access to historical crime data, socioeconomic data, and real-time incident reports. Which method will help in determining the data SMEs for this project?

The most effective method for identifying data Subject Matter Experts (SMEs) is to find individuals with experience on similar projects. For a predictive policing AI tool, this experience implies a deep understanding of the specific data types involved (historical crime, socioeconomic, real-time incidents), their inherent biases, quality issues, and contextual relevance. This holistic, domain-specific knowledge is crucial for guiding data preparation, feature engineering, and model validation, which is more valuable than narrow expertise in one data type or general technical certifications.

A. This method is too narrow, as it only assesses knowledge of real-time incident data, ignoring the critical historical and socioeconomic data components.
C. This is also too narrow, as it omits the real-time data aspect and may unnecessarily limit the search to the existing project team.
D. This identifies technical skills in data science, not subject matter expertise in policing, crime data, or socioeconomics, which is essential for ensuring data quality and relevance.

1. Project Management Institute (PMI). While the CPMAI is a new certification, foundational PMI principles in the PMBOK® Guide – Seventh Edition emphasize the role of the project team and stakeholders. Section 4.3, "Project Team," notes the importance of "individuals with specialized knowledge or skills" relevant to the project's context. For an AI project in policing, prior experience on similar tools is the most direct indicator of this specialized knowledge.

2. Arkhipova, K., & Faria, P. (2021). The Role of the Project Manager in Artificial Intelligence Projects. In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS), Vol. 2, pages 715-722. DOI: 10.5220/0010490907150722. This paper highlights that AI projects require a blend of skills, and the project manager must secure "domain experts" who understand the business context and data intricacies. Identifying individuals from similar past projects is a direct way to secure this expertise.

3. Lum, K., & Isaac, W. (2016). To predict and serve? Significance, 13(5), 14-19. DOI: 10.1111/j.1740-9713.2016.00960.x. This academic review on predictive policing discusses the profound impact of data selection and interpretation. It implicitly supports the need for SMEs with deep contextual understanding of crime and socioeconomic data, a level of expertise most likely found in individuals who have previously worked on such complex, high-stakes systems.