Fundamentals Of Neural Networks Architectures Algorithms And Applications

McCabe's cyclomatic complexity metric (1976) is widely cited as a useful predictor of various software attributes such as reliability and development effort This critique demonstrates that it is based upon poor theoretical foundations and an inadequate model of software development The argument that the metric provides the developer with a useful engineering approximation is not borne out by the empirical evidence Furthermore, it would appear that for a large class of software it is no more than a proxy for, and in many cases is outperformed by, lines of code<
 >

A critique of cyclomatic complexity as a software metric

Multi-sensor information fusion based on machine learning for real applications in human activity recognition: State-of-the-art and research challenges

Human Activity Recognition (HAR) employing inertial motion data has gained considerable momentum in recent years, both in research and industrial applications. From the abstract perspective, this has been driven by an acceleration in the building of intelligent and smart environments and systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc. Such environments and systems necessitate and subsume activity recognition, aimed at recognizing the actions, characteristics, and goals of one or more individuals from a temporal series of observations streamed from one or more sensors. Due to the reliance of conventional Machine Learning (ML) techniques on handcrafted features in the extraction process, current research suggests that deep-learning approaches are more applicable to automated feature extraction from raw sensor data. In this work, the generic HAR framework for smartphone sensor data is proposed, based on Long Short-Term Memory (LSTM) networks for time-series domains. Four baseline LSTM networks are comparatively studied to analyze the impact of using different kinds of smartphone sensor data. In addition, a hybrid LSTM network called 4-layer CNN-LSTM is proposed to improve recognition performance. The HAR method is evaluated on a public smartphone-based dataset of UCI-HAR through various combinations of sample generation processes (OW and NOW) and validation protocols (10-fold and LOSO cross validation). Moreover, Bayesian optimization techniques are used in this study since they are advantageous for tuning the hyperparameters of each LSTM network. The experimental results indicate that the proposed 4-layer CNN-LSTM network performs well in activity recognition, enhancing the average accuracy by up to 2.24% compared to prior state-of-the-art approaches.

https://www.mdpi.com/1424-8220/21/5/1636/pdf

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

Over the last decade, speech emotion recognition (SER) has become an interesting and challenging topic in the human behavior analysis research field. The objective of this area of research is to classify the emotional states of people based on the speech patterns of humans. Currently, the focus of this research field is on the identification of the effectiveness of automatic classifiers that can enhance the efficiency of the classification in practical applications, particularly those used in telecommunication services. Negative emotions, such as sadness, anger, disgust, and fear, can provide a significant amount of beneficial data to both the user of the quality assurance platform and the customer. This paper examines the complicated task involving recognition of negative emotions in human speech data by employing a deep learning technique. Four open emotional speech datasets are used in this research in order to identify a deep learning classifier that provides good efficiency for use with negative emotion speech data. Furthermore, the classifier with the best performance was also tested with a Thai language dataset. Based on the experimental results, the one-dimensional convolution neural network was determined to be the classifier that has the most outstanding performance level for tasks involving negative emotion recognition in the Thai language with a level of accuracy at 96.60%.

Negative Emotion Recognition using Deep Learning for Thai Language

Wearable technology has advanced significantly and is now used in various entertainment and business contexts. Authentication methods could be trustworthy, transparent, and non-intrusive to guarantee that users can engage in online communications without consequences. An authentication system on a security framework starts with a process for identifying the user to ensure that the user is permitted. Establishing and verifying an individual’s appearance usually requires a lot of effort. Recent years have seen an increase in the usage of activity-based user identification systems to identify individuals. Despite this, there has not been much research into how complex hand movements can be used to determine the identity of an individual. This research used a one-dimensional residual network with squeeze-and-excitation (SE) configurations called the 1D-ResNet-SE model to investigate hand movements and user identification. According to the findings, the SE modules have enhanced the one-dimensional residual network’s identification ability. As a deep learning model, the proposed methodology is capable of effectively identifying features from the input smartwatch sensor and could be utilized as an end-to-end model to clarify the modeling process. The 1D-ResNet-SE identification model is superior to the other models. Hand movement assessment based on deep learning is an effective technique to identify smartwatch users.

https://www.mdpi.com/1424-8220/22/8/3094/pdf?version=1650354593

Deep Residual Network for Smartwatch-Based User Identification through Complex Hand Movements

Numerous learning-based techniques for effective human behavior identification have emerged in recent years. These techniques focus only on fundamental human activities, excluding transitional activities due to their infrequent occurrence and short period. Nevertheless, postural transitions play a critical role in implementing a system for recognizing human activity and cannot be ignored. This study aims to present a hybrid deep residual model for transitional activity recognition utilizing signal data from wearable sensors. The developed model enhances the ResNet model with hybrid Squeeze-and-Excitation (SE) residual blocks combining a Bidirectional Gated Recurrent Unit (BiGRU) to extract deep spatio-temporal features hierarchically, and to distinguish transitional activities efficiently. To evaluate recognition performance, the experiments are conducted on two public benchmark datasets (HAPT and MobiAct v2.0). The proposed hybrid approach achieved classification accuracies of 98.03% and 98.92% for the HAPT and MobiAct v2.0 datasets, respectively. Moreover, the outcomes show that the proposed method is superior to the state-of-the-art methods in terms of overall accuracy. To analyze the improvement, we have investigated the effects of combining SE modules and BiGRUs into the deep residual network. The findings indicates that the SE module is efficient in improving transitional activity recognition.

/pdf/a-hybrid-deep-residual-network-for-efficient-transitional-14xylyel.pdf

A Hybrid Deep Residual Network for Efficient Transitional Activity Recognition Based on Wearable Sensors

Recognition of human behavior is recently an active and stimulating study. The HAR can provide valuable information on human movement and the behavior of everyday life activities. In the last decade, a large range of HAR-based applications have been implemented, such as healthcare tracking, biometric user authentication, and so on. Previously, several deep learning approaches have been introduced to focus on the issue of conventional machine learning approaches with handcrafted features. So, a novel deep learning architecture to solve the HAR problem is proposed in this study. The introduced architecture is a hybrid model called a multichannel CNN-LSTM network. The model is evaluated by state-of-the-art evaluation metrics; accuracy, precision, recall and F1-score, with a public dataset of smartwatch’s accelerometer data called DHA dataset. The proposed multichannel CNN-LSTM outperforms other deep learning methods in terms of accuracy, with a score of 96.87%.

A Multichannel CNN-LSTM Network for Daily Activity Recognition using Smartwatch Sensor Data

Due to its vast applications in various industrial sectors, sensor-based human activity recognition (SHAR) has become a prevalent study issue in machine learning (ML) and deep learning (DL). With the improvement of numerous wearable sensors, many effective use cases have recently been revealed. According to recent research, real-world data contains more contextual information than data acquired in a laboratory environment. Three deep learning models were used to investigate real-life activities using smartphone sensors in this study. As two fundamental deep learning approaches, a convolutional neural network (CNN) and a long short-term memory (LSTM) network are used to achieve recognition. In addition, we introduced the Att-CNN-LSTM network as a hybrid DL model to handle the SHAR challenge using an attention mechanism. On a public dataset called real-life HAR (RL-HAR), these three deep learning models were evaluated using four assessment indicators: accuracy, precision, recall, and F1-score. According to the experimental data, the suggested Att-CNN-LSTM surpasses existing baseline deep learning models with the highest average accuracy of 95.76%.

Sakorn Mekruksavanich

Papers

Negative Emotion Recognition using Deep Learning for Thai Language

Deep Residual Network for Smartwatch-Based User Identification through Complex Hand Movements

A Hybrid Deep Residual Network for Efficient Transitional Activity Recognition Based on Wearable Sensors

A Multichannel CNN-LSTM Network for Daily Activity Recognition using Smartwatch Sensor Data

Recognition of Real-life Activities with Smartphone Sensors using Deep Learning Approaches