SEGMENTED–MEMORY RECURRENT NEURAL NETWORKS VERSUS HIDDEN MARKOV MODELS IN EMOTION RECOGNITION FROM SPEECH

AbstractThe ability to correctly predict the functional role of proteins from their amino acid sequences would significantly advance biological studies at the molecular level by improving our ability to understand the biochemical capability of biological organisms from their genomic sequence. Existing methods that are geared towards protein function prediction or annotation mostly use alignment-based approaches and probabilistic models such as Hidden-Markov Models. In this work we introduce a deep learning architecture (FunctionIdentification withNeuralDescriptions orFIND) which performs protein annotation from primary sequence. The accuracy of our methods matches state of the art techniques, such as protein classifiers based on Hidden Markov Models. Further, our approach allows for model introspection via a neural attention mechanism, which weights parts of the amino acid sequence proportionally to their relevance for functional assignment. In this way, the attention weights automatically uncover structurally and functionally relevant features of the classified protein and find novel functional motifs in previously uncharacterized proteins. While this model is applicable to any database of proteins, we chose to apply this model to superfamilies of homologous proteins, with the aim of extracting features inherent to divergent protein families within a larger superfamily. This provided insight into the functional diversification of an enzyme superfamily and its adaptation to different physiological contexts. We tested our approach on three families (nitrogenases, cytochromebd-type oxygen reductases and heme-copper oxygen reductases) and present a detailed analysis of the sequence characteristics identified in previously characterized proteins in the heme-copper oxygen reductase (HCO) superfamily. These are correlated with their catalytic relevance and evolutionary history. FIND was then applied to discover features in previously uncharacterized members of the HCO superfamily, providing insight into their unique sequence features. This modeling approach demonstrates the power of neural networks to recognize patterns in large datasets and can be utilized to discover biochemically and structurally important features in proteins from their amino acid sequences.Author summary

Download Full-text

Comparison Between Hidden Markov Models and Artificial Neural Networks in the Classification of Bearing Defects

Applied Condition Monitoring - Rotating Machinery and Signal Processing ◽

10.1007/978-3-319-96181-1_6 ◽

2018 ◽

pp. 68-78

Author(s):

Miloud Sedira ◽

Ridha Ziani ◽

Ahmed Felkaoui

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Hidden Markov Models ◽

Markov Models ◽

Hidden Markov ◽

Bearing Defects ◽

Artificial Neural

Download Full-text

A New Modular Strategy For Action Sequence Automation Using Neural Networks And Hidden Markov Models

International Journal of System Dynamics Applications ◽

10.4018/ijsda.2013070102 ◽

2013 ◽

Vol 2 (3) ◽

pp. 18-35 ◽

Cited By ~ 5

Author(s):

Mohamed Adel Taher ◽

Mostapha Abdeljawad

Keyword(s):

Neural Networks ◽

Hidden Markov Models ◽

Fuzzy Systems ◽

Markov Models ◽

Hidden Markov ◽

Current System ◽

Action Sequence ◽

Hybrid Strategy ◽

Automatic Control Theory ◽

Welding Automation

In this paper, the authors propose a new hybrid strategy (using artificial neural networks and hidden Markov models) for skill automation. The strategy is based on the concept of using an “adaptive desired” that is introduced in the paper. The authors explain how using an adaptive desired can help a system for which an explicit model is not available or is difficult to obtain to smartly cope with environmental disturbances without requiring explicit rules specification (as with fuzzy systems). At the same time, unlike the currently available hidden Markov-based systems, the system does not merely replay a memorized skill. Instead, it takes into account the current system state as reported by sensors. The authors approach can be considered a bridge between the spirit of conventional automatic control theory and fuzzy/hidden Markov-based thinking. To demonstrate the different aspects of the proposed strategy, the authors discuss its application to underwater welding automation.

Download Full-text