Advanced Signal Processing and Adaptive Learning Methods
Zoran Perić, Vlado Delić, Zoran Stamenković, David D. Pokrajac · Computational Intelligence and Neuroscience · 2019
Research in the areas of signal processing and artificial intelligence (and developed methods and algorithms) has become increasingly important in the last two decades.In this special issue, we present new ideas and hybrid approaches based on techniques from the aforementioned scientific areas encouraging researchers from different fields to adopt them in accomplishing complex multidisciplinary tasks. is special issue includes a set of novel contributions, which covers a wide range of advanced signal processing techniques and adaptive learning methods for various engineering purposes.In the paper entitled "RnRTD: Intelligent Approach Based on the Relationship-Driven Neural Network and Restricted Tensor Decomposition for Multiple Accusation Judgment in Legal Cases," X. Guo et al. describe a new method for judging multiple crimes in legal cases.is is a multilabel classification, and it is based on the relationshipdriven recurrent neural network (rdRNN) and restricted tensor decomposition (RTD).e authors have demonstrated that the proposed RTD layer and the relation-driven cyclic neural network have remarkable optimization effects on various deep neural network algorithms.Next, a novel emotion identification method based on mutual information feature weight, which captures the correlation and redundancy of features, is presented in the paper entitled "Adaptive Learning Emotion Identification Method of Short Texts for Online Medical Knowledge Sharing Community" by D. Gan et al.e paper entitled "Speech Technology Progress Based on New Machine Learning Paradigm" by V. Delić et al. provides an overview of speech technologies development as