Safe: Machine unlearning with shard graphs

Y Dukler, B Bowman, A Achille… - Proceedings of the …, 2023 - openaccess.thecvf.com
Abstract We present Synergy Aware Forgetting Ensemble (SAFE), a method to adapt large
models on a diverse collection of data while minimizing the expected cost to remove the …

AI model disgorgement: Methods and choices

A Achille, M Kearns, C Klingenberg… - Proceedings of the …, 2024 - pnas.org
Over the past few years, machine learning models have significantly increased in size and
complexity, especially in the area of generative AI such as large language models. These …

A new hazard event classification model via deep learning and multifractal

Z Wang, B Wang, M Ren, D Gao - Computers in Industry, 2023 - Elsevier
Hazard and operability analysis (HAZOP) is the paradigm of industrial safety that can reveal
the hazards of process from its node deviations, consequences, causes, measures and …

Linking negative herding effect to online public opinion: A new BiGADG approach for sentiment classification

Y Zhou, Y Hu, H Liu, P Huang, D Dai - Expert Systems, 2024 - Wiley Online Library
The negative herding effect tends to cause the spread of negative emotions among netizens,
resulting in significant public opinion crises. Therefore, linking the negative herding effect …

GRALA: modeling social information for microblog sentiment analysis from the view of balancing sparsity and smoothness of social contexts

X Zou, S Hu, T Li - 2023 Asia Pacific Signal and Information …, 2023 - ieeexplore.ieee.org
As online social networks provide people with a popular and convenient platform to express
opinions, microblog sentiment analysis has become a hot research area. Currently, social …

A Multi-dimension and Multi-granularity Feature Fusion Method for Chinese Microblog Sentiment Classification

T Wei, X Liu, L Qu, Y Chen - Proceedings of the 2023 6th International …, 2023 - dl.acm.org
Chinese microblog comments exhibit a wide range of expressions, including brief terms and
internet slang. Traditional single-feature models may not capturing this diversity …

Building Classes for Sentiment of Social Media Content Using WU Semantic Measure

A Kayed, MA Saidi - The International Conference on Innovations in …, 2022 - Springer
Sentiment classification or opinion mining is conducted to extract the core data representing
opinions, and emotions. Motivated by the role of social media on appointing the dimension …