Energy-based Domain Adaption with Active Learning for Emerging Misinformation Detection

Lee, Kyumin; Mou, Guanyi; Sievert, Scott

doi:10.1109/bigdata55660.2022.10021038

Cited by 2 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most probabilistic models can be considered as special types of energy-based models, where the energy function satisfies certain normalization conditions and the loss function is optimized by learning and has a specific form [3]. We could leverage the energy-based learning method to detect outliers within a data set [4][5].In this paper, we need to consider many factors, develop ESM simulation models that can be applied to many types of ESM through mathematical modeling, construct a simulation algorithm to simulate the real situation, and on this basis, analyze the sensitivity of the model using new impact indicators, and extend the model to be applied to the research mission [6][7].…”

Section: Introductionmentioning

confidence: 99%

Research and application based on simulation of ESM model

Wang,

Fu,

Zhang

2023

Third International Conference on Advanced Algorithms and Signal Image Processing (AASIP 2023)

View full text Add to dashboard Cite

In this paper, simulation is based on the ESM model, which is traditionally characterized by high nonlinearity and high uncertainty. The complexity of the system poses a great challenge for its overall characterization. Modern simulation technology can perform detailed numerical modeling and simulation calculations to realize the analysis and prediction of the system operation state, providing an effective research tool and decision support for the construction of the ESM model optimization system. In this paper, the simulation application of ESM model is based on matrix equation and dynamical system, and the sensitivity analysis is attempted, which proves that the simulation effect of ESM model is good and has good research value and application significance, in order to provide some reference significance for the research in other fields.

show abstract

Section: Introductionmentioning

confidence: 99%

Research and application based on simulation of ESM model

Wang,

Fu,

Zhang

2023

Third International Conference on Advanced Algorithms and Signal Image Processing (AASIP 2023)

View full text Add to dashboard Cite

show abstract

Towards Data- and Compute-Efficient Fake-News Detection: An Approach Combining Active Learning and Pre-Trained Language Models

Folino,

Guarascio

et al. 2024

SN COMPUT. SCI.

View full text Add to dashboard Cite

In today’s digital era, dominated by social media platforms such as Twitter, Facebook, and Instagram, the swift dissemination of misinformation represents a significant concern, impacting public sentiment and influencing pivotal global events. Promptly detecting such deceptive content with the help of Machine Learning models is crucial, yet it comes with the challenge of dealing with labelled examples for training these models. Impressive performance results were recently achieved by high-capacity pre-trained transformer-based models (e.g., BERT). Still, such models are too data- and compute-demanding for many critical application contexts where memory, time, and energy consumption must be limited. Here, we propose an innovative semi-supervised method for efficient and effective fake news detection using a content-oriented classifier based on a small-sized BERT embedder. After fine-tuning this model on the sole few labelled data available, an iterative Active Learning (AL) process is carried out, which benefits from limited experts’ feedback to acquire more labelled data for improving the model. The proposed method ensures good detection performances using a few training samples, reasonably small human intervention, and compute/memory costs.

show abstract

Energy-based Domain Adaption with Active Learning for Emerging Misinformation Detection

Cited by 2 publications

References 9 publications

Research and application based on simulation of ESM model

Research and application based on simulation of ESM model

Towards Data- and Compute-Efficient Fake-News Detection: An Approach Combining Active Learning and Pre-Trained Language Models

Contact Info

Product

Resources

About