A Multi-attention Collaborative Deep Learning Approach for Blood Pressure Prediction

We develop a deep learning model based on Long Short-term Memory (LSTM) to predict blood pressure based on a unique data set collected from physical examination centers capturing comprehensive multi-year physical examination and lab results. In the Multi-attention Collaborative Deep Learning model (MAC-LSTM) we developed for this type of data, we incorporate three types of attention to generate more explainable and accurate results. In addition, we leverage information from similar users to enhance the predictive power of the model due to the challenges with short examination history. Our model significantly reduces predictive errors compared to several state-of-the-art baseline models. Experimental results not only demonstrate our model’s superiority but also provide us with new insights about factors influencing blood pressure. Our data is collected in a natural setting instead of a setting designed specifically to study blood pressure, and the physical examination items used to predict blood pressure are common items included in regular physical examinations for all the users. Therefore, our blood pressure prediction results can be easily used in an alert system for patients and doctors to plan prevention or intervention. The same approach can be used to predict other health-related indexes such as BMI.

Automatic missing value imputation for cleaning phase of diabetic’s readmission prediction model

Recently, the industry of healthcare started generating a large volume of datasets. If hospitals can employ the data, they could easily predict the outcomes and provide better treatments at early stages with low cost. Here, data analytics (DA) was used to make correct decisions through proper analysis and prediction. However, inappropriate data may lead to flawed analysis and thus yield unacceptable conclusions. Hence, transforming the improper data from the entire data set into useful data is essential. Machine learning (ML) technique was used to overcome the issues due to incomplete data. A new architecture, automatic missing value imputation (AMVI) was developed to predict missing values in the dataset, including data sampling and feature selection. Four prediction models (i.e., logistic regression, support vector machine (SVM), AdaBoost, and random forest algorithms) were selected from the well-known classification. The complete AMVI architecture performance was evaluated using a structured data set obtained from the UCI repository. Accuracy of around 90% was achieved. It was also confirmed from cross-validation that the trained ML model is suitable and not over-fitted. This trained model is developed based on the dataset, which is not dependent on a specific environment. It will train and obtain the outperformed model depending on the data available.

Determining customer limits by data mining methods in credit allocation process

The demand for credit is increasing constantly. Banks are looking for various methods of credit evaluation that provide the most accurate results in a shorter period in order to minimize their rising risks. This study focuses on various methods that enable the banks to increase their asset quality without market loss regarding the credit allocation process. These methods enable the automatic evaluation of loan applications in line with the sector practices, and enable determination of credit policies/strategies based on actual needs. Within the scope of this study, the relationship between the predetermined attributes and the credit limit outputs are analyzed by using a sample data set of consumer loans. Random forest (RF), sequential minimal optimization (SMO), PART, decision table (DT), J48, multilayer perceptron(MP), JRip, naïve Bayes (NB), one rule (OneR) and zero rule (ZeroR) algorithms were used in this process. As a result of this analysis, SMO, PART and random forest algorithms are the top three approaches for determining customer credit limits.

A Comparative Study of Energy Big Data Analysis for Product Management in a Smart Factory

Energy has been obtained as one of the key inputs for a country's economic growth and social development. Analysis and modeling of industrial energy are currently a time-insertion process because more and more energy is consumed for economic growth in a smart factory. This study aims to present and analyse the predictive models of the data-driven system to be used by appliances and find out the most significant product item. With repeated cross-validation, three statistical models were trained and tested in a test set: 1) General Linear Regression Model (GLM), 2) Support Vector Machine (SVM), and 3) boosting Tree (BT). The performance of prediction models measured by R2 error, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Variation (CV). The best model from the study is the Support Vector Machine (SVM) that has been able to provide R2 of 0.86 for the training data set and 0.85 for the testing data set with a low coefficient of variation, and the most significant product of this smart factory is Skelp.

Application of improved convolution neural network in financial forecasting

Financial status and its role in the national economy have been increasingly recognized. In order to deduce the source of monetary funds and determine their whereabouts, financial information and prediction have become a scientific method that can not be ignored in the development of national economy. This paper improves the existing CNN and applies it to financial credit from different perspectives. Firstly, the noise of the collected data set is deleted, and then the clustering result is more stable by principal component analysis. The observation vectors are segmented to obtain a set of observation vectors corresponding to each hidden state. Based on the output of PCA algorithm, we recalculate the mean and variance of all kinds of observation vectors, and use the new mean and covariance matrix as credit financial credit, and then determine the best model parameters.The empirical results based on specific data from China's stock market show that the improved convolutional neural network proposed in this paper has advantages and the prediction accuracy reaches.

Hybridizing Convolutional Neural Network for Classification of Lung diseases

Pulmonary disease is widespread worldwide. There is persistent blockage of the lungs, pneumonia, asthma, TB, etc. It is essential to diagnose the lungs promptly. For this reason, machine learning models were developed. For lung disease prediction, many deep learning technologies, including the CNN, and the capsule network, are used. The fundamental CNN has low rotating, inclined, or other irregular image orientation efficiency. Therefore by integrating the space transformer network (STN) with CNN, we propose a new hybrid deep learning architecture named STNCNN. The new model is implemented on the dataset from the Kaggle repository for an NIH chest X-ray image. STNCNN has an accuracy of 69% in respect of the entire dataset, while the accuracy values of vanilla grey, vanilla RGB, hybrid CNN are 67.8%, 69.5%, and 63.8%, respectively. When the sample data set is applied, STNCNN takes much less time to train at the cost of a slightly less reliable validation. Therefore both specialists and physicians are simplified by the proposed STNCNN System for the diagnosis of lung disease.

The challenges of emotion recognition methods based on electroencephalogram signals: a literature review

Electroencephalogram (EEG) signals in recognizing emotions have several advantages. Still, the success of this study, however, is strongly influenced by: i) the distribution of the data used, ii) consider of differences in participant characteristics, and iii) consider the characteristics of the EEG signals. In response to these issues, this study will examine three important points that affect the success of emotion recognition packaged in several research questions: i) What factors need to be considered to generate and distribute EEG data?, ii) How can EEG signals be generated with consideration of differences in participant characteristics?, and iii) How do EEG signals with characteristics exist among its features for emotion recognition? The results, therefore, indicate some important challenges to be studied further in EEG signals-based emotion recognition research. These include i) determine robust methods for imbalanced EEG signals data, ii) determine the appropriate smoothing method to eliminate disturbances on the baseline signals, iii) determine the best baseline reduction methods to reduce the differences in the characteristics of the participants on the EEG signals, iv) determine the robust architecture of the capsule network method to overcome the loss of knowledge information and apply it in more diverse data set.

Malware Detection in Android Apps Using Static Analysis

Frequency of malware attacks because Android apps are increasing day by day. Current studies have revealed startling facts about data harvesting incidents, where user’s personal data is at stake. To preserve privacy of users, a permission induced risk interface MalApp to identify privacy violations rising from granting permissions during app installation is proposed. It comprises of multi-fold process that performs static analysis based on app’s category. First, concept of reverse engineering is applied to extract app permissions to construct a Boolean-valued permission matrix. Second, ranking of permissions is done to identify the risky permissions across category. Third, machine learning and ensembling techniques have been incorporated to test the efficacy of the proposed approach on a data set of 404 benign and 409 malicious apps. The empirical studies have identified that our proposed algorithm gives a best case malware detection rate of 98.33%. The highlight of interface is that any app can be classified as benign or malicious even before running it using static analysis.

Malware Detection in Android Apps Using Static Analysis

Frequency of malware attacks because Android apps are increasing day by day. Current studies have revealed startling facts about data harvesting incidents, where user’s personal data is at stake. To preserve privacy of users, a permission induced risk interface MalApp to identify privacy violations rising from granting permissions during app installation is proposed. It comprises of multi-fold process that performs static analysis based on app’s category. First, concept of reverse engineering is applied to extract app permissions to construct a Boolean-valued permission matrix. Second, ranking of permissions is done to identify the risky permissions across category. Third, machine learning and ensembling techniques have been incorporated to test the efficacy of the proposed approach on a data set of 404 benign and 409 malicious apps. The empirical studies have identified that our proposed algorithm gives a best case malware detection rate of 98.33%. The highlight of interface is that any app can be classified as benign or malicious even before running it using static analysis.