LSSVM-based social spam detection model
Xiaolei Yang, SU Yi-dan, Jin Ping Mo · Proceedings of the 2nd International Conference On Systems Engineering and Modeling · 2013
To Resolve the garbage tag issue in Folksonomy, Lssvm algorithm for social spa m detection model (least Squares support vector machine classifiers) was proposed.The met hod of inequality change the constraints in the traditional support vector machine into equa lity constraints, and take the empirical function of the squared error loss function as the E xperience function in training set.so that the quadratic programming problem convert QP i nto solving linear equations, it was improving solution the speed of solution and accuracy o f convergence.The experimental results show that we have got higher classification accuracy and less predict time than traditional svm detection methods based on least squares support vector machine algorithm garbage tag detection model.