Existing Machine Learning Algorithms Research Articles

An Efficient Prediction Method for Coronary Heart Disease Risk Based on Two Deep Neural Networks Trained on Well-Ordered Training Datasets

This study proposes an efficient prediction method for coronary heart disease risk based on two deep neural networks trained on well-ordered training datasets. Most real datasets include an irregular subset with higher variance than most data, and predictive models do not learn well from these datasets. While most existing prediction models learned from the whole or randomly sampled training datasets, our suggested method draws up training datasets by separating regular and highly biased subsets to build accurate prediction models. We use a two-step approach to prepare the training dataset: (1) divide the initial training dataset into two groups, commonly distributed and highly biased using Principal Component Analysis, (2) enrich the highly biased group by Variational Autoencoders. Then, two deep neural network classifiers learn from the isolated training groups separately. The well-organized training groups enable a chance to build more accurate prediction models. When predicting the risk of coronary heart disease from the given input, only one appropriate model is selected based on the reconstruction error on the Principal Component Analysis model. Dataset used in this study was collected from the Korean National Health and Nutritional Examination Survey. We have conducted two types of experiments on the dataset. The first one proved how Principal Component Analysis and Variational Autoencoder models of the proposed method improves the performance of a single deep neural network. The second experiment compared the proposed method with existing machine learning algorithms, including Naive Bayes, Random Forest, K-Nearest Neighbor, Decision Tree, Support Vector Machine, and Adaptive Boosting. The experimental results show that the proposed method outperformed conventional machine learning algorithms by giving the accuracy of 0.892, specificity of 0.840, precision of 0.911, recall of 0.920, f-measure of 0.915, and AUC of 0.882.

P110 A soft, flexible, wearable device for cough detection in paediatric cystic fibrosis patients

Objectives: New or increased cough is an important sign of pulmonary exacerbation. There is unmet need to objectively quantify and analyse cough trends in early life. Advances in flexible electronics and materials allow development of soft, skin-like, accelerometer-based wearable devices that seamlessly interface with the human body in unique locations, enabling the simultaneous capture of low-frequency body, chest, and throat motion along with high-frequency vocal, throat, and lung sound signals associated with various body processes including coughs and vital signs. Methods: A small, flexible, fully wireless, accelerometer-based mechano-acoustic sensor (MAS) was applied using gentle adhesives to the suprasternal notch of 10 paediatric CF subjects during regular clinic visits. A 15-minute protocol consisting of eliciting cough, throat clearing, speech, and laughs in various head/body orientations, ambient environments, and physical activity intensities. The MAS recorded at a high bandwidth 1.6kHz rate and automatically uploaded to a cloud server upon replacement on a wireless charging platform. Results: The captured sensor data was analysed using an existing machine learning algorithm developed for COVID-19 symptom tracking. Preliminary results show the device differentiates cough from other vocal, respiratory noises and motion artifacts, both at rest and during activity. A majority of children found the device to be acceptable, though device size and adhesive removal caused minor discomfort for some. Conclusion: The extraction of well-classified cough events is feasible, facilitating further analysis of cough episode duration, force, and clinically relevant features associated with early decompensation, response to treatment, and daily symptom tracking. Objective measurement of cough may be useful as a clinical study outcome measure and for clinical monitoring. Future work will include evaluation for longer time periods during stability and pulmonary exacerbations.

A Quantum Spatial Graph Convolutional Network for Text Classification

The data generated from non-Euclidean domains and its graphical representation (with complex-relationship object interdependence) applications has observed an exponential growth. The sophistication of graph data has posed consequential obstacles to the existing machine learning algorithms. In this study, we have considered a revamped version of a semi-supervised learning algorithm for graph-structured data to address the issue of expanding deep learning approaches to represent the graph data. Additionally, the quantum information theory has been applied through Graph Neural Networks (GNNs) to generate Riemannian metrics in closed-form of several graph layers. In further, to pre-process the adjacency matrix of graphs, a new formulation is established to incorporate high order proximities. The proposed scheme has shown outstanding improvements to overcome the deficiencies in Graph Convolutional Network (GCN), particularly, the information loss and imprecise information representation with acceptable computational overhead. Moreover, the proposed Quantum Graph Convolutional Network (QGCN) has significantly strengthened the GCN on semi-supervised node classification tasks. In parallel, it expands the generalization process with a significant difference by making small random perturbations of the graph during the training process. The evaluation results are provided on three benchmark datasets, including Citeseer, Cora, and PubMed, that distinctly delineate the superiority of the proposed model in terms of computational accuracy against state-of-the-art GCN and three other methods based on the same algorithms in the existing literature.

DeepPPF: A deep learning framework for predicting protein family

Machine learning pipelines for protein functional family prediction are urgently needed especially now that only 1% of raw protein sequences have been manually annotated. Although existing machine learning algorithms have achieved a decent performance in modeling and predicting the functional families of protein sequences, they still have two drawbacks. First, biological dependencies among nucleotides are not rich enough to describe motifs for these methods. Also, existing algorithms are not accurate enough to predict the functional families of newly discovered proteins. To address the above limitations simultaneously, we propose a novel deep learning framework for predicting protein family, DeepPPF, which employs the word2vec technique in capturing distributional dependencies among nucleotides and discovers rich features from diverse motif lengths to characterize proteins. The novelty of the DeepPPF is in utilizing distributional dependencies among nucleotides. Experimental results on G protein-coupled receptor hierarchical datasets show the effectiveness of DeepPPF in achieving the state-of-the-art performance in items of Mathew’s correlation coefficients (MCC) of 97.62%, 88.45% and, 83.09% for family, sub-family and, sub-subfamily hierarchical levels, respectively. Also, DeepPPF outperformed existing methods in terms of prediction accuracy and Mathew’s correlation coefficients on the cluster of orthologous groups (COG) and phage of orthologous groups (POG) datasets. Furthermore, we analyzed the ability of DeepPPF framework to discover rich motifs for functional classes with the least sets of protein sequences. The experimental results show that rich motif discovery is key to improving the modeling performance of protein families through deep learning techniques. Finally, we investigated the effect of transferring a low-level functional domain level to a high-level functional domain and results show that the target domain prediction can be improved with transfer learning. Therefore, our proposed deep learning framework can be useful in characterizing protein functional families. The codes and datasets are available at https://github.com/CSUBioGroup/DeepPPF.

Analysis on intelligent machine learning enabled with meta-heuristic algorithms for solar irradiance prediction

The solar forecasting is an effective method to enhance the operation of an electrical system for merging a large amount of solar power generation systems and intends to expand a new empirical method to model the prediction uncertainty of the solar irradiance. The proposed model comprises three phases, such as (a) Data Acquisition, (b) Feature Extraction, and (c) Prediction. Initially, benchmark data available from local meteorological organizations are collected that includes the numerical weather forecasting data like temperature, dew point, humidity, visibility, wind speed, and other descriptive information. Once the data is collected, feature extraction is done by first-order and second-order statistical models. First Order Statistics, like mean, median, standard deviation, the maximum value of entire data, and minimum value of entire data, and Second-Order Statistics, like Kurtosis, skewness, correlation, and entropy are extracted as the features. These features are further applied to three machine learning algorithms named Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN). As main novelty of this paper, the number of hidden neurons of all these networks is optimized by a hybrid algorithm merging both the Deer Hunting Optimization Algorithm (DHOA) and Grey Wolf Optimization (GWO), which is named as Grey Updated DHOA (GU-DHOA). The improvement of these networks with the assistance of a hybrid meta-heuristic algorithm will be highly effective for solar irradiance prediction, overcoming the existing machine learning algorithms.

A new feature selection using dynamic interaction

With the continuous development of Internet technology, data gradually present a complicated and high-dimensional trend. These high-dimensional data have a large number of redundant features and irrelevant features, which bring great challenges to the existing machine learning algorithms. Feature selection is one of the important research topics in the fields of machine learning, pattern recognition and data mining, and it is also an important means in the data preprocessing stage. Feature selection is to look for the optimal feature subset from the original feature set, which would improve the classification accuracy and reduce the machine learning time. The traditional feature selection algorithm tends to ignore the kind of feature which has a weak distinguishing capacity as a monomer, whereas the feature group’s distinguishing capacity is strong. Therefore, a new dynamic interaction feature selection (DIFS) algorithm is proposed in this paper. Initially, under the theoretical framework of interactive information, it redefines the relevance, irrelevance and redundancy of the features. Secondly, it offers the computational formulas for calculating interactive information. Finally, under the eleven data sets of UCI and three different classifiers, namely, KNN, SVM and C4.5, the DIFS algorithm increases the classification accuracy of the FullSet by 3.2848% and averagely decreases the number of features selected by 15.137. Hence, the DIFS algorithm can not only identify the relevance feature effectively, but also identify the irrelevant and redundant features. Moreover, it can effectively improve the classification accuracy of the data sets and reduce the feature dimensions of the data sets.

DeTrAs: deep learning-based healthcare framework for IoT-based assistance of Alzheimer patients

AbstractHealthcare 4.0 paradigm aims at realization of data-driven and patient-centric health systems wherein advanced sensors can be deployed to provide personalized assistance. Hence, extreme mentally affected patients from diseases like Alzheimer can be assisted using sophisticated algorithms and enabling technologies. Motivated from this fact, in this paper, DeTrAs: Deep Learning-based Internet of Health Framework for the Assistance of Alzheimer Patients is proposed. DeTrAs works in three phases: (1) A recurrent neural network-based Alzheimer prediction scheme is proposed which uses sensory movement data, (2) an ensemble approach for abnormality tracking for Alzheimer patients is designed which comprises two parts: (a) convolutional neural network-based emotion detection scheme and (b) timestamp window-based natural language processing scheme, and (3) an IoT-based assistance mechanism for the Alzheimer patients is also presented. The evaluation of DeTrAs depicts almost 10–20% improvement in terms of accuracy in contrast to the different existing machine learning algorithms.

Distributed learning machines for solving forward and inverse problems in partial differential equations

We conceptualize Distributed Learning Machines (DLMs) – a novel machine learning approach that integrates existing machine learning algorithms with traditional mesh-based numerical methods for solving forward and inverse problems in nonlinear partial differential equations (PDEs). In conventional numerical methods such as finite element method (FEM), the discretization of the computational domain is a standard technique to reduce the representation load of basis functions. Along the same lines, we propose a distributed neural network architecture that facilitates the simultaneous deployment of several localized neural networks to solve PDEs in a unified manner. The most critical requirement of the DLMs is the synchronization of the distributed neural networks. For this, we introduce a new physics-based interface regularization term to the cost function of the existing learning machines like the Physics Informed Neural Network (PINN) and the Physics Informed Extreme Learning Machine (PIELM). To evaluate the efficacy of this approach, we develop three distinct variants of DLM namely, time-marching Distributed PIELM (DPIELM), Distributed PINN (DPINN) and time-marching DPINN. We show that ideas of linearization and time-marching allow DPIELM to be able to solve nonlinear PDEs to some extent. Next, we show that DPINNs have potential advantages over existing PINNs to solve the inverse problems in heterogeneous media. Finally, we propose a rapid, time-marching version of DPINN which leverages the ideas of transfer learning to accelerate the training. Collectively, this framework leads towards the promise of hybrid Neural Network-FVM or Neural Network-FEM schemes in the future.

An Efficient Intrusion Detection Method Based on LightGBM and Autoencoder

Due to the insidious characteristics of network intrusion behaviors, developing an efficient intrusion detection system is still a big challenge, especially in the era of big data where the number of traffic and the dimension of each traffic feature are high. Because of the shortcomings of traditional common machine learning algorithms in network intrusion detection, such as insufficient accuracy, a network intrusion detection system based on LightGBM and autoencoder (AE) is proposed. The LightGBM-AE model proposed in this paper includes three steps: data preprocessing, feature selection, and classification. The LightGBM-AE model adopts the LightGBM algorithm for feature selection, and then uses an autoencoder for training and detection. When a set of data containing network intrusion behaviors are inputted into an autoencoder, there is a large reconstruction error between the original input data and the reconstructed data obtained by the autoencoder, which provides a basis for intrusion detection. According to the reconstruction error, an appropriate threshold is set to distinguish symmetrically between normal behavior and attack behavior. The experiment is carried out on the NSL-KDD dataset and implemented using Pytorch. In addition to autoencoder, variational autoencoder (VAE) and denoising autoencoder (DAE) are also used for intrusion detection and are compared with existing machine learning algorithms such as Decision Tree, Random Forest, KNN, GBDT, and XGBoost. The evaluation is carried out through classification evaluation indexes such as accuracy, precision, recall, F1-score. The experimental results show that the method can efficiently separate the attack behavior from normal behavior according to the reconstruction error. Compared with other methods, the effectiveness and superiority of this method are verified.

Classification of Sea Ice Types in Sentinel-1 SAR Data Using Convolutional Neural Networks

A new algorithm for classification of sea ice types on Sentinel-1 Synthetic Aperture Radar (SAR) data using a convolutional neural network (CNN) is presented. The CNN is trained on reference ice charts produced by human experts and compared with an existing machine learning algorithm based on texture features and random forest classifier. The CNN is trained on two datasets in 2018 and 2020 for retrieval of four classes: ice free, young ice, first-year ice and old ice. The accuracy of our classification is 90.5% for the 2018-dataset and 91.6% for the 2020-dataset. The uncertainty is a bit higher for young ice (85%/76% accuracy in 2018/2020) and first-year ice (86%/84% accuracy in 2018/2020). Our algorithm outperforms the existing random forest product for each ice type. It has also proved to be more efficient in computing time and less sensitive to the noise in SAR data. The code is publicly available.

A Two-layer Mechanism Identification Method for VIN Adversarial Examples

With the rapid development of machine learning algorithms, the security problems are gradually emerging. Most existing machine learning algorithms may be attacked by adversarial examples. Especially in the domain of path planning, the adversarial maps may result in multiple harmful consequences on the paths predicted by Deep Reinforcement Learning (DRL) algorithms. However, there is no suitable approach to automatically identify them. To our knowledge, all previous work used manual observation method to identify the adversarial maps, which is time-consuming. Therefore, this paper explores a two-layer mechanism method to automatically identify the adversarial examples in Value Iteration Networks (VIN). We define the four categories of attack results and identify them by combining path feature comparison and path image classification. Experiments show that this method can achieve an effective identification on VIN adversarial examples.

Lbl Lstm Log Bilinear And Long Short Term Memory Based Efficient Stock Forecasting Model Considering External Fluctuating Factor

Stock market prediction problem is considered to be NP-hard problem because of highly volatile nature of stock market. In this paper, effort has been made to design efficient stock forecasting model using log Bilinear and long short term memory (LBL-LSTM) considering external fluctuating factor such as varying Bank's lending rates. The external factor bank's lending rates affects stock market performance ,as it plays vital role for the purchase of stocks in case of financial crisis faced by various business enterprises. Proposed LBL-LSTM based model shows performance improvement over existing machine learning algorithms used for stock market prediction.

Big Data Classification and Internet of Things in Healthcare

Every single day, a massive amount of data is generated by different medical data sources. Processing this wealth of data is indeed a daunting task, and it forces us to adopt smart and scalable computational strategies, including machine intelligence, big data analytics, and data classification. The authors can use the Big Data analysis for effective decision making in healthcare domain using the existing machine learning algorithms with some modification to it. The fundamental purpose of this article is to summarize the role of Big Data analysis in healthcare, and to provide a comprehensive analysis of the various techniques involved in mining big data. This article provides an overview of Big Data, applicability of it in healthcare, some of the work in progress and a future works. Therefore, in this article, the use of machine learning techniques is proposed for real-time diabetic patient data analysis from IoT devices and gateways.

Soft Clustering for Enhancing the Diagnosis of Chronic Diseases over Machine Learning Algorithms.

Chronic diseases represent a serious threat to public health across the world. It is estimated at about 60% of all deaths worldwide and approximately 43% of the global burden of chronic diseases. Thus, the analysis of the healthcare data has helped health officials, patients, and healthcare communities to perform early detection for those diseases. Extracting the patterns from healthcare data has helped the healthcare communities to obtain complete medical data for the purpose of diagnosis. The objective of the present research work is presented to improve the surveillance detection system for chronic diseases, which is used for the protection of people's lives. For this purpose, the proposed system has been developed to enhance the detection of chronic disease by using machine learning algorithms. The standard data related to chronic diseases have been collected from various worldwide resources. In healthcare data, special chronic diseases include ambiguous objects of the class. Therefore, the presence of ambiguous objects shows the availability of traits involving two or more classes, which reduces the accuracy of the machine learning algorithms. The novelty of the current research work lies in the assumption that demonstrates the noncrisp Rough K-means (RKM) clustering for figuring out the ambiguity in chronic disease dataset to improve the performance of the system. The RKM algorithm has clustered data into two sets, namely, the upper approximation and lower approximation. The objects belonging to the upper approximation are favourable objects, whereas the ones belonging to the lower approximation are excluded and identified as ambiguous. These ambiguous objects have been excluded to improve the machine learning algorithms. The machine learning algorithms, namely, naïve Bayes (NB), support vector machine (SVM), K-nearest neighbors (KNN), and random forest tree, are presented and compared. The chronic disease data are obtained from the machine learning repository and Kaggle to test and evaluate the proposed model. The experimental results demonstrate that the proposed system is successfully employed for the diagnosis of chronic diseases. The proposed model achieved the best results with naive Bayes with RKM for the classification of diabetic disease (80.55%), whereas SVM with RKM for the classification of kidney disease achieved 100% and SVM with RKM for the classification of cancer disease achieved 97.53 with respect to accuracy metric. The performance measures, such as accuracy, sensitivity, specificity, precision, and F-score, are employed to evaluate the performance of the proposed system. Furthermore, evaluation and comparison of the proposed system with the existing machine learning algorithms are presented. Finally, the proposed system has enhanced the performance of machine learning algorithms.

Zero-Shot Human Activity Recognition Using Non-Visual Sensors.

Due to significant advances in sensor technology, studies towards activity recognition have gained interest and maturity in the last few years. Existing machine learning algorithms have demonstrated promising results by classifying activities whose instances have been already seen during training. Activity recognition methods based on real-life settings should cover a growing number of activities in various domains, whereby a significant part of instances will not be present in the training data set. However, to cover all possible activities in advance is a complex and expensive task. Concretely, we need a method that can extend the learning model to detect unseen activities without prior knowledge regarding sensor readings about those previously unseen activities. In this paper, we introduce an approach to leverage sensor data in discovering new unseen activities which were not present in the training set. We show that sensor readings can lead to promising results for zero-shot learning, whereby the necessary knowledge can be transferred from seen to unseen activities by using semantic similarity. The evaluation conducted on two data sets extracted from the well-known CASAS datasets show that the proposed zero-shot learning approach achieves a high performance in recognizing unseen (i.e., not present in the training dataset) new activities.

An Online Calculator for the Prediction of Survival in Glioblastoma Patients Using Classical Statistics and Machine Learning.

BACKGROUNDAlthough survival statistics in patients with glioblastoma multiforme (GBM) are well-defined at the group level, predicting individual patient survival remains challenging because of significant variation within strata.OBJECTIVETo compare statistical and machine learning algorithms in their ability to predict survival in GBM patients and deploy the best performing model as an online survival calculator.METHODSPatients undergoing an operation for a histopathologically confirmed GBM were extracted from the Surveillance Epidemiology and End Results (SEER) database (2005-2015) and split into a training and hold-out test set in an 80/20 ratio. Fifteen statistical and machine learning algorithms were trained based on 13 demographic, socioeconomic, clinical, and radiographic features to predict overall survival, 1-yr survival status, and compute personalized survival curves.RESULTSIn total, 20 821 patients met our inclusion criteria. The accelerated failure time model demonstrated superior performance in terms of discrimination (concordance index = 0.70), calibration, interpretability, predictive applicability, and computational efficiency compared to Cox proportional hazards regression and other machine learning algorithms. This model was deployed through a free, publicly available software interface (https://cnoc-bwh.shinyapps.io/gbmsurvivalpredictor/).CONCLUSIONThe development and deployment of survival prediction tools require a multimodal assessment rather than a single metric comparison. This study provides a framework for the development of prediction tools in cancer patients, as well as an online survival calculator for patients with GBM. Future efforts should improve the interpretability, predictive applicability, and computational efficiency of existing machine learning algorithms, increase the granularity of population-based registries, and externally validate the proposed prediction tool.

Predicting bipolar disorder and schizophrenia based on non-overlapping genetic phenotypes using deep neural network

Computational Psychiatry is an emerging field of science. It focuses on identifying the complex relationship between the brain’s neurobiology. Mental illness has recently become an important problem to be addressed as the number of people affected is increasing over time. Schizophrenia and Bipolar Disorder are two major types of psychiatric disorders. Most of the people are experienced these illness in their lifetime. But, diagnosing psychiatric disorders is even more a complex problem. Genetic factors play a vital role in developing mental illness. Interestingly, few psychiatric disorders have common genetic overlapping between each other. It causes detrimental effect on diagnosing the illness accurately. To overcome this existing issue, a Rank based Gene Biomarker Identification and Classification framework is proposed to identify the overlapping and non-overlapping gene patterns of bipolar disorder and schizophrenia. The dataset used in this experiment is obtained from Gene Expression Omnibus database. As an outcome of this experiment, seven biomarkers are identified as the overlapping genes. Also, 60 and 68 informative gene biomarkers are identified on bipolar disorder and schizophrenia dataset as feature subsets to discriminate the samples. Overlapping genes are eliminated to increase the diagnostic accuracy of the disorders. The performance of the proposed system is evaluated with standard existing machine learning algorithms. This proposed framework attained 97.01% and 95.65% accuracy on bipolar disorder and schizophrenia dataset with Deep Neural Network model outperformed other benchmarked algorithms and proved its efficacy.

Water scarcity prediction for global region using machine learning

Water is a big challenge not only in India but in many countries of the world. Machine learning and forecasting model has been employed towards water demand and ground water level prediction. But in terms of water scarcity, much less work has been carried out by employing machine learning algorithms like 'artificial neural network' (ANN) and 'grey forecasting' model for forecasting water scarcity and none has focused on historical data like water availability, water consumption for a particular area and stress value for predicting water scarcity. So accordingly, we here have developed a water scarcity prediction system based on historical data by employing 'deep neural networks' which is an advanced form of 'artificial neural networks'. We have also compared 'deep neural network' with existing machine learning algorithms such as "support vector machine (SVM), logistic regression and Naive Bayes". From the analysis of algorithms based on dataset, deep neural networks have been found as the best prediction model for water scarcity.

XGBoost-PCA-BPNN prediction model and its application on predicting the effectiveness of non-surgical periodontal treatment

An XGBoost-PCA-BPNN classification and predication model for the big data set with uncertain and coupling multidimensional characteristics is presented to solve the problem that the existing machine learning algorithms cannot get the high prediction performances. The clinical data set of the NSPT for Chinese population with CP is a typical big data set with uncertain and coupling multidimensional characteristics. The XGBoost-PCA-BPNN model is constructed to predict the effectiveness of NSPT for Chinese population with CP in this paper. The model is verified by applying it to predict the effectiveness of NSTP based on 45,000 clinical sites data set with eight characteristics. Prediction results show that the R2-score obtained by applying the Xgboost-PCA-BPNN model is 0.943, which is higher than those obtained by using Logistic regression, Xgboost, and Xgboost-Logistic regression, respectively. In addition, it is found that the effectiveness of NSTP is mainly influenced by PD, BI, sites and age.

Water scarcity prediction for global region using machine learning

Water is a big challenge not only in India but in many countries of the world. Machine learning and forecasting model has been employed towards water demand and ground water level prediction. But in terms of water scarcity, much less work has been carried out by employing machine learning algorithms like 'artificial neural network' (ANN) and 'grey forecasting' model for forecasting water scarcity and none has focused on historical data like water availability, water consumption for a particular area and stress value for predicting water scarcity. So accordingly, we here have developed a water scarcity prediction system based on historical data by employing 'deep neural networks' which is an advanced form of 'artificial neural networks'. We have also compared 'deep neural network' with existing machine learning algorithms such as support vector machine (SVM), logistic regression and Naive Bayes. From the analysis of algorithms based on dataset, deep neural networks have been found as the best prediction model for water scarcity.