ML

Large Language Models and Personal Information: Security Challenges and Solutions Through Anonymization

ctive methods to protect personal data in online texts. Existing anonymization methods often prove ineffective against complex LLM analysis algorithms, especially when processing sensitive information such as medical data. This research proposes an innovative approach to anonymization that combines k-anonymity and adversarial methods. Our approach aims to improve the efficiency and speed of anonymization while maintaining a high level of data protection.

Evaluating machine learning models efficacy in sentiment analysis for Moroccan Darija: An exploration with MAC dataset

Sentiment analysis is an essential technique for classifying and extracting emotions from several data sets.  While many basic methods distinguish between negative and positive emotions, advanced approaches may consider additional categories, such as neutral emotions.  This becomes very important and difficult when we need to deal with less parsed languages and dialects, such as Moroccan Darija.  Our study highlights the nuances of conducting sentiment analysis implementing the MAC dataset, which includes comments in Moroccan Darija.  Our main target is to do comparativ