This talk will explore the concept of automated text anonymization powered by machine learning, a solution that can quickly and accurately protect sensitive data. The talk will cover all the necessary concepts and considerations to implement such a solution in the industry. Some of the key concepts that will be discussed include the data labeling process, training NLP models, a human-in-the-loop process to ensure privacy and monitor the performance of the ML model, and implementation of MLOPS principles.

Target Audience: ML Engineers, Data Scientists, Decision-Makers in the area of data privacy
Prerequisites: Familiarity with basic machine learning methods
Level: Advanced

Extended Abstract:
Data privacy and anonymization are important topics and will probably stay in the future. Balancing data privacy and data utilization is a crucial challenge for many organizations. On one hand, companies must protect sensitive information and comply with regulations. On the other hand, they can unlock new business insights from the text data they collect. This talk will explore the concept of automated text anonymization powered by machine learning, a solution that effectively balances both considerations by quickly and accurately protecting sensitive data while preserving its utility. The talk will cover all the necessary concepts and considerations to implement such a solution in the industry. Some of the key concepts that will be discussed include the data labeling process, training NLP models, a human-in-the-loop process to ensure privacy and monitor the performance of the ML model, and MLOPS concepts to ensure the model improves over time