Enhancing Workforce Intelligence: How msg Utilizes AI-Based Data Harmonization with Amazon Bedrock

Co-authored by Stefan Walter from msg

This title captures the essence of the collaboration and highlights the focus on AI and data harmonization in the context of workforce management.

Transforming HR with AI: The Story of msg.ProfileMap

This post is co-written with Stefan Walter from msg.

Introduction

In a world increasingly driven by data, the need for organizations to harness their human resource (HR) information has never been more critical. msg, a global leader with over 10,000 experts in 34 countries, has been paving the way in this arena for over 40 years. Their innovative solution, msg.ProfileMap, is a Software as a Service (SaaS) offering that is revolutionizing skill and competency management in HR departments. With more than 7,500 users and strong compliance with industry standards, msg.ProfileMap enables organizations to effectively map workforce capabilities, identify skill gaps, and implement targeted development strategies.

In this post, we’ll explore how msg automated data harmonization for msg.ProfileMap, utilizing Amazon Bedrock to streamline data enrichment workflows, resulting in better accuracy, reduced manual labor, and enhanced compliance with the EU AI Act and GDPR.

The Importance of AI-Based Data Harmonization

HR departments often grapple with managing inconsistent and fragmented data. Critical documents might be unstructured, while legacy systems rely on outdated formats. This lack of standardization not only jeopardizes data quality but also hinders decision-making. Accurate and harmonized HR data is essential for activities such as matching candidates to roles, identifying internal mobility opportunities, and conducting skills gap analysis.

msg recognized that without scalable methods to process and unify HR data, organizations would continue to suffer from manual overhead and inconsistent outcomes.

Solution Overview

HR data typically comes from various sources, including relational databases, Excel files, and Word documents, each with different identifiers and descriptions. To tackle this challenge, msg developed a modular architecture focused on IT workforce scenarios.

At the core of msg.ProfileMap lies a robust text extraction layer that converts diverse inputs into structured data. This structured data is sent to an AI-powered harmonization engine designed to ensure data consistency across sources, avoiding duplication and aligning disparate concepts.

The harmonization process employs a hybrid retrieval approach that utilizes both vector-based semantic similarity and string-based matching techniques. With Amazon Bedrock’s capabilities, msg semantically enriches data, resulting in improved compatibility and precision, which is then indexed using Amazon OpenSearch Service and Amazon DynamoDB for rapid retrieval.

This unsupervised framework, optimized for IT workflows, also shows strong capabilities across various domains, expanding its usefulness beyond the HR sector.

Results and Technical Validation

To measure the effectiveness of this data harmonization framework, msg conducted internal tests and participated in the Bio-ML Track of the Ontology Alignment Evaluation Initiative (OAEI).

Internal testing demonstrated that the system processed 2,248 concepts, achieving an impressive 95.5% accuracy in high-probability merge recommendations, covering nearly 60% of inputs. This resulted in a over 70% reduction in manual validation workload—a significant advantage for HR teams.

In the OAEI 2024 benchmark, msg.ProfileMap excelled, scoring top marks across multiple biomedical datasets, including a remarkable 0.918 F1 score on NCIT-DOID.

Why Amazon Bedrock?

The choice of Amazon Bedrock was crucial for msg in leveraging Large Language Models (LLMs) for near real-time data enrichment. With its fully managed, serverless interface, Amazon Bedrock simplifies scaling and operational demands, allowing msg to focus on delivering value rather than managing infrastructure.

This model aligns seamlessly with msg’s SaaS delivery strategy and compliance needs—especially when handling sensitive HR data under the EU AI Act and GDPR—by ensuring auditable interactions with model APIs.

Conclusion

The integration of Amazon Bedrock into msg.ProfileMap illustrates how large-scale AI adoption can be achieved without complex infrastructure. By combining a modular design with ontology-based harmonization and the capabilities of Amazon Bedrock, msg has crafted a workforce intelligence platform that is accurate, scalable, and compliant.

This solution not only enhances concept matching precision but also ranks highly in international benchmarks, showcasing the potential of generative AI when paired with advanced cloud services. With msg.ProfileMap, organizations are fiercely equipped to meet today’s HR challenges and anticipate tomorrow’s demands.

msg.ProfileMap is available as a SaaS offering on AWS Marketplace. If you are interested in learning more, please reach out to msg at msg.hcm.backoffice@msg.group.

About the Authors

Stefan Walter is Senior Vice President of AI SaaS Solutions at msg. With over 25 years of IT experience, he leads innovative initiatives that merge business strategy with technology execution.

Gianluca Vegetti is a Senior Enterprise Architect in the AWS Partner Organization, specializing in Strategic Collaboration Agreements with AWS partners.

Yuriy Bezsonov is a Senior Partner Solution Architect at AWS, aiding partners and customers with cloud solutions and focusing on container technologies, application modernization, and Generative AI.

The content and views expressed in this post are those of the authors, with AWS not responsible for the content’s accuracy.