Deep semi‐supervised classification based in deep clustering and cross‐entropy

Abstract

Self‐labeled techniques, a semi‐supervised classification paradigm (SSC), are highly effective in alleviating the scarcity of labeled data used in classification tasks through an iterative process of self‐training. This problem was addressed by several approaches with different assumptions about the features of the input data, examples of these approaches being self‐training, co‐training, STRED, among others. This paper presents a framework for data self‐labeling based on deep autoencoder combined with a self‐labeled technique that takes into consideration cross‐entropy. The model uses the Encoder to reduce the dimensionality of the input that is submitted to a labeling layer. The weights of this layer are adjusted through the learning from a clustering performed in the Z space, which is the reduced dimensionality space. Results showed that the proposed method obtained competitive performance in relation to classic methods that are found in the literature.

Assessing cloud QoS predictions using OWA in neural network methods

“…Different gradient descent optimisation techniques such as RMSProp, Adamax [ 13 ], and Adam are commonly used to address computational complexity problems. Deep autoencoder [ 14 ] is used to reduce the dimensionality of the input in a labelling layer [ 15 , 16 ]. Most machine learning methods have convergence issues towards the global minimum.…”

Predictive intelligence using ANFIS‐induced OWAWA for complex stock market prediction

“…31,32 Gated recurrent units (GRUs) and long short-term memory (LSTM) are some of the widely used recurrent neural network (RNN) methods for different time series predictions. 33,34 Different variations in parameters and combinations of various hybrid approaches are tested to improve the prediction accuracy. de Lima et al 33 used a deep NN model utilising stock data and transaction records to predict future intervals.…”

“…33,34 Different variations in parameters and combinations of various hybrid approaches are tested to improve the prediction accuracy. de Lima et al 33 used a deep NN model utilising stock data and transaction records to predict future intervals. The analysis results revealed that bidirectional LSTM could accurately forecast stock prices for the economic selection process.…”

“…Deep learning methods have simplified supervised learning and tend to produce optimised outcomes 31,32 . Gated recurrent units (GRUs) and long short‐term memory (LSTM) are some of the widely used recurrent neural network (RNN) methods for different time series predictions 33,34 . Different variations in parameters and combinations of various hybrid approaches are tested to improve the prediction accuracy.…”