Rumors Detection Based on Lifelong Machine Learning

Abstract

The amount of training data is small in the field of Weibo rumor detection, and online news changes constantly, but the existing models do not have the ability of continuous learning, which means they cannot achieve knowledge accumulation and update. Hence, they will require many training examples to improve their detection ability. In contrast, Lifelong Machine Learning (LML) paradigm has the capability of continuous learning, which retains the knowledge learned in the past and uses it to help future learning tasks. After learning, some knowledge is updated. With the growth and update of knowledge, the performance of each task will be better and better. Hence, we use this paradigm to build a Weibo rumor detection model. First, we extracted three types of features based on content, user, and propagation from Weibo events and proposed three new propagation features and Bidirectional Encoder Representations from Transformers (BERT) semantic features of source message for rumor detection. We then used Simulated Annealing (SA) to improve the Genetic Algorithm (GA), which was called GA-SA used to search for the best global minimum feature subset to enhance the classification effect of the Efficient Lifelong Learning Algorithm (ELLA) on rumors. The ELLA transfers knowledge to learn new tasks and refines knowledge over time to maximize performance across all tasks in the continuous learning process. The proposed model is called GA-SA-ELLA. The experimental results show that our model could achieve superior detection results even with less training data for each task.