Machine Learning Environment Research Articles

P.35 Genetic newborn screening and digital technologies to accelerate rare disease diagnosis: The EU-IMI screen4Care project

Between 5000 and 8000 rare diseases (RDs) affect the daily lives of around 30 million people in the EU, representing a significant challenge for the healthcare system and society. RDs patients often experience long and burdensome path to diagnosis, possible misdiagnosis, or improper treatment, which may be due to low prevalence of RDs, high number of different diseases and the lack of general sensibleness. It is estimated that 72% of RDs are genetic and 70% have an exclusively paediatric onset. Strategies for genetic newborn screening (NBS) are thus of pivotal importance, since an early RDs diagnosis would radically change their clinical history. Screen4Care (S4C), a new EU Research Project, is a public-private collaboration of 35 partners led by the University of Ferrara and including 21 academic partners, 9 industrial project partners -led by Pfizer-, 4 small and medium-sized enterprises and EURORDIS. S4C -that will last five years- offers an innovative research approach to accelerate rare disease diagnosis, through two central pillars: genetic NBS and artificial intelligence (AI)-based tools such as machine learning. The first central pillar, genetic NBS, will adopt different strategies interrogating: I) currently treatable RDs (TREAT-panel-approach), II) actionable RDs (ACT-panel-approach); III) and whole genome sequencing (WGS) will be offered to symptomatic patients when NBS resulted negative. Moreover, S4C aims to use the power of innovative digital solutions to shorten time to diagnosis via two routes: I) predictive algorithms, leveraging the Screen4Care data (federated) machine learning environment and embedded electronic health record (EHR) systems; II) a repository of AI "symptom checkers", which will be designed to facilitate self-diagnosis and/or to suggest referral pathways to physicians for diagnostic workup. S4C aims to shorten the path to RDs diagnosis with this dual approach and to establish a digital infrastructure for families and health workers.

The Influence of Psychological Distance on the Challenging Moral Decision Support of Sports Majors in Internet of Things and Machine Learning

This work intends to examine the influence of different dimensions of psychological distance on the moral decision-making of sports college students in sports dilemmas under different learning pressure conditions, and to further investigate the relationship between psychological distance and moral decision-making. The research on the influencing factors of moral decision-making of sports majors can effectively help to understand the moral cognition level of the group, and provide a reference for the interpretation of athletes’ moral anomie behavior, thereby enriching the content of the moral quality education of athletes. This work intends to study the impact of psychological distance on the moral decision support of sports college students in the context of the Internet of Things (IoT) and machine learning. Psychological distance in the machine learning environment may affect individuals’ understanding and cognition of events and, to a certain extent, can change students’ cognition and judgment of events. IoT and machine learning environments are chosen as the foundation. The learning pressure of college students majoring in physical education is a variable. A questionnaire survey and experimental design are used to test the influence of different degrees of learning pressure, social distance, and spatial distance on the moral decision-making of physical education college students in the sports dilemma. The dimensions of the psychological distance of physical education (PE) students are analyzed under different stress conditions and their impact on the moral decision-making of PE students. This experiment adopts a mixed experimental design of 3 learning stresses (no stress vs. moderate stress vs. high stress) × 2 social distances (self vs. others) × 2 spatial distances (Beijing vs. France). The results show that the main effect of social distance is significant. When the self is the decision-making subject, individuals tend to make more moral decisions. There is a significant interaction between social distance and learning pressure. In a stress-free and high-stress environment, individuals make a significant increase in the number of moral decisions when faced with self-centered decision-making. Now, moral decision-making and its consequences are important for college students majoring in sports. The results of their moral decision-making in the field of education directly reflect the image of the individual and even the institution.

Open problems in medical federated learning

PurposeThis study aims to summarize the critical issues in medical federated learning and applicable solutions. Also, detailed explanations of how federated learning techniques can be applied to the medical field are presented. About 80 reference studies described in the field were reviewed, and the federated learning framework currently being developed by the research team is provided. This paper will help researchers to build an actual medical federated learning environment.Design/methodology/approachSince machine learning techniques emerged, more efficient analysis was possible with a large amount of data. However, data regulations have been tightened worldwide, and the usage of centralized machine learning methods has become almost infeasible. Federated learning techniques have been introduced as a solution. Even with its powerful structural advantages, there still exist unsolved challenges in federated learning in a real medical data environment. This paper aims to summarize those by category and presents possible solutions.FindingsThis paper provides four critical categorized issues to be aware of when applying the federated learning technique to the actual medical data environment, then provides general guidelines for building a federated learning environment as a solution.Originality/valueExisting studies have dealt with issues such as heterogeneity problems in the federated learning environment itself, but those were lacking on how these issues incur problems in actual working tasks. Therefore, this paper helps researchers understand the federated learning issues through examples of actual medical machine learning environments.

Network Data Center Traffic Predictive Model Analysis Based on Machine Learning

Due to the transient nature and uncertainty of traffic produced by applications and services, data center networks have a lot of challenges. As a response, networking as a domain is continually evolving to maintain the exponential growth in network traffic. The primary objective of this paper is to predict the network traffic before it impacts the system's performance. This paper first describes existing Machine Learning (ML) applications in telecommunications and then lists the most prominent difficulties and probable remedies for implementing them. We tried to implement different ML algorithms to predict the network traffic like Gradient Boosting (GB), Random Forest (RF), K-Nearest Neighbor (KNN), Adaptive Boosting (AB), Neural Network (NN), Decision Tree (DT), and Support Vector Machines (SVM) with different sub-parameters for predicting network traffic. Relying on a sequential dataset, we create the corresponding ML environment and present a comparison table of Mean square error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Determination (R2) for each model. The simulation results show that the AB and GB are the best-fitted models with performance matrix parameters like MSE 0.000 and RMSE 0.002 and 0.011, respectively. The orange tool is used to stimulate the predictive models.

SYNTHESIS OF THE REWARD FUNCTION IN REINFORCED LEARNING WITH COGNITIVE GRAPHICS

In recent years, reinforcement learning methods have become increasingly important for many applied areas. Such learning assumes the presence of a reward function. The reward should be the higher, the more the agent’s behavior corresponds to the desired one. At the same time, in many cases, the reward function is built axiomatically, by expert selection of one of the most widely used functions based on a superficial analysis of the subject area. This situation is due to cognitive difficulties encountered by an expert in the process of constructing reward functions, especially for a large number of arguments. At the same time, the reward function can be represented as an aggregation operator, since the range of valid values of any criterion can be reduced to a unit interval by an appropriate linear transformation. Thus, the task of constructing a reward function can be reduced to the task of constructing an aggregation operator with given properties. To ensure the intuitive clarity of the process of constructing aggregation operators, a method for their visualization using 3D-cognitive graphics has been developed. This article proposes a method for synthesizing the reward function for reinforcement learning, which includes the mentioned visualization. The synthesis method includes two procedures, which are sequences of steps performed by an expert, each of which requires him to take specific actions. An experiment was set up to test the effectiveness of the developed method. During this experiment, the synthesis of the reward function and reinforcement learning of agents based on the synthesized function were implemented in the multiagent machine learning environment of the StarCraft II computer game. Training was conducted for the standard StarCraft II reward function and for the reward function built using the proposed procedure.

Machine Learning Based Network Traffic Predictive Analysis

Digital enterprises that use various Internet of Things implementation prototypes, such as cloud, mobile, and edge equipment, are experiencing unprecedented traffic volume and dynamicity. Data center networks (DCN) have faced various issues due to the transient and random nature of traffic created by services and apps. The primary objective of this paper is to predict the network traffic using the machine learning (ML) models before the performance of the network start degrading. Because in the last decade, ML has had a tremendous impact on handling the massive amount of data. With the increase in complexity and traffic, we tried to implement four ML models such as K – Nearest Neighbor (KNN), Random Forest (RF), Gradient Boosting (GB), and Decision Tree (Tree), with tuned sub-parameters to predict the network traffic. We create a matching ML environment based on a sequential database and provide a comparison table of mean square error (MSE), root mean square error (RMSE), mean absolute error (MSE), and coefficient of determination (R2) for each prototype. The simulation results show that the GB with different types is the best-suited model for predicting the network traffic with performance matrix parameters such as MSE 0.001 and RMSE of 0.030. Therefore, the Orange tool is used to stimulate the predictive models.

Numerical study of two heat exchangers for the cooling of a battery pack for an electric vehicle

Background: The thermal management of a battery pack designed for an electric vehicle is a key to prevent accidental events and ensure a long lifespan of the batteries. A typical accident is a thermal runaway of one or more cells in the battery which can cause fire or explosion of the battery pack. This paper presents a numerical modelling of a battery pack (BP) and its heat exchanger (HE) for an electric vehicle. The heat produced in the battery is evacuated by the HE. Methods: Two different kinds of modelling have been realized: a computational fluid dynamic (CFD) modelling and a coarse (called MOD3 for 3D Model) modelling. The CFD modelling allows the creation of fine numerical simulations of a BP, but uses large meshes, therefore the cost of each calculation is important. In order to make a large number of quite long transient simulations, a second tool called MOD3, employing only a coarse mesh, was developed in this study at the Commissariat à l’énergie atomique et aux énergies alternatives (CEA). Results: Two measurement campaigns corresponding to two different versions of the HE have been conducted at CEA. The temperature measurements allow comparisons of MOD3 to a real battery pack and to fit some heat exchange coefficients. The cells temperatures as well as the cooling liquid temperature are compared. Conclusions: The MOD3 tool has been fitted partly on CFD calculations, and partly on experimental measurements. It will be integrated in a machine learning environment by CIDETEC to take into account the thermal management of the BP in real car simulations.

Visual Design of Brand Image by Marine Bionic System in the Environment of Internet of Things and Machine Learning

Based on the environment of economic globalization, many clothing companies have increased the requirements for clothing design styles to diversify them, thereby increasing the interest of consumers. The main purpose and motivation are to explore the pattern design of children’s clothing with the marine bionic environment as the design source. Firstly, the collaborative business concept of IoT machine learning is introduced into the field of clothing design. Design styles and elements related to marine bionic environments are introduced. A set of questionnaires about the visual design preferences of children in different families on marine style clothing patterns are designed to examine differences in the cognitive level of color patterns among children of different age groups. Through the sorting, statistics, and analysis of the questionnaire results, different families have a stronger interest in clothing with warm colors as the color style and lines and ordinary paintings as the pattern drawing style. This provides a certain degree of design ideas for related clothing design work and provides unique insights into the visual design of brand image based on the marine bionic system in the Internet of Things (IoT) and machine learning environment.

Implementation of Artificial intelligence for maintenance operation in the rail industry

The quest to enhance maintenance operation in the rail industry has continued to occupy the front burner in recent time in a bid to reduce machine’s downtime and ensure smooth operations. In this study, Artificial Intelligence (AI) technique was proposed for maintenance operation of wheel-bearing component of a railcar. Data from a secondary source was pre-processed and iteratively trained using specialised training algorithm in a machine learning environment under a supervised training until it produces a model capable of making predictions. The result obtained indicates the feasibility of the developed AI model for prediction of the Remaining Useful Life of a wheel-bearing component. The result shows that the wheel-bearing will last for 500 hours over the next 40 days before it begins to fail in service. The wheel bearing starts showing sign of degradation on day 41 of usage. Upon the use of the predictive model, the predicted RUL, confidence bound and slope detection instant were obtained. Hence, the implementation of AI for predictive maintenance could promote maintenance operation in the rail industry.

Boosting Climate Analysis With Semantically Uplifted Knowledge Graphs

Nowadays, the fast expansion of heterogeneous climate data resources accessible on the Internet has led to substantial data fragmentation on the web. For example, station-based sensor data from different sources are likely to be interrelated but may be stored in disparate formats, such as <monospace>CSV</monospace>, <monospace>JSON</monospace>, and <monospace>XML</monospace>. To address the data isolation problem, several semantically uplifted knowledge graphs are proposed for climate data exchange. While these knowledge graphs improve data interoperability, the advancement in multisource data interchange is limited to data included inside knowledge graphs. As a result, the exclusive interoperability of current climatic knowledge graphs hampers the flow of data into typical climate analysis workflows in contexts, where analytical models often need data in nonknowledge graph formats. This article addresses this issue by focusing on enhancing climate analysis workflows within the context of the Python machine learning environment, with a preference for tabular data. We propose an analysis workflow able to automatically integrate remote climate knowledge graph data with local tabular data so as to enhance the data usability with respect to certain climate analysis tasks. To underscore the importance of our study, we illustrate how the workflow streamlines the access to multisource climatic variables in the Python environment from a semantic perspective. The additional knowledge graph data have the potential to augment local datasets in the climate domain, as evidenced by an improvement in accuracy of up to 10&#x0025; for machine learning geared on rainfall detection.

Disease detection and consultation using django and machine learning

The design and implementation of various well-known data mining techniques in a variety of real-world applications (e.g., industry, healthcare, and bioscience) has led to their use in machine learning environments to extract meaningful information from provided data in healthcare communities, biological disciplines, and other fields. Early illness prediction, patient treatment, and community services all benefit from precise medical database analysis. Machine learning techniques have been effectively used in a variety of applications, including disease prediction. The goal of developing a classifier system utilising machine learning algorithms is to greatly assist physicians in predicting and diagnosing diseases at an early stage, which will greatly aid in solving health-related difficulties. For our study, a sample of 4920 patient records diagnosed with 41 disorders was chosen. We chose 95 out of 132 independent variables (symptoms) that are strongly associated to illnesses and improved them. The disease prediction system built utilising Machine learning techniques such as Decision Tree classifier, Random forest classifier, and Nave Bayes classifier is demonstrated in this research paper. This paper “Disease Prediction Using Django and Machine Learning” gives a comparison of the outcomes of the aforementioned algorithms.

Mapping high resolution National Soil Information Grids of China

Soil spatial information has traditionally been presented as polygon maps at coarse scales. Solving global and local issues, including food security, water regulation, land degradation, and climate change requires higher quality, more consistent and detailed soil information. Accurate prediction of soil variation over large and complex areas with limited samples remains a challenge, which is especially significant for China due to its vast land area which contains the most diverse soil landscapes in the world. Here, we integrated predictive soil mapping paradigm with adaptive depth function fitting, state-of-the-art ensemble machine learning and high-resolution soil-forming environment characterization in a high-performance parallel computing environment to generate 90-m resolution national gridded maps of nine soil properties (pH, organic carbon, nitrogen, phosphorus, potassium, cation exchange capacity, bulk density, coarse fragments, and thickness) at multiple depths across China. This was based on approximately 5000 representative soil profiles collected in a recent national soil survey and a suite of detailed covariates to characterize soil-forming environments. The predictive accuracy ranged from very good to moderate (Model Efficiency Coefficients from 0.71 to 0.36) at 0–5 cm. The predictive accuracy for most soil properties declined with depth. Compared with previous soil maps, we achieved significantly more detailed and accurate predictions which could well represent soil variations across the territory and are a significant contribution to the GlobalSoilMap.net project. The relative importance of soil-forming factors in the predictions varied by specific soil property and depth, suggesting the complexity and non-stationarity of comprehensive multi-factor interactions in the process of soil development.

OpenNERO: A Game Platform for AI Research and Education

OpenNERO is an open source game platform designed for game AI research. The software package combines features commonly available in modern game engines (such as 3D graphics, physics simulation, 3D audio rendering, networked play, and a powerful scripting interface) with an easy to use API and tools for defining machine learning tasks, environments, and agents. Flexibility and ease of use of the system are demonstrated by following the process of creating a machine learning game from scratch. The scalability of the platform is tested through the implementation of the existing NERO machine learning game using the new tools.

Posits and the state of numerical representations in the age of exascale and edge computing

AbstractGrowing constraints on memory utilization, power consumption, and I/O throughput have increasingly become limiting factors to the advancement of high performance computing (HPC) and edge computing applications. IEEE‐754 floating‐point types have been the de facto standard for floating‐point number systems for decades, but the drawbacks of this numerical representation leave much to be desired. Alternative representations are gaining traction, both in HPC and machine learning environments. Posits have recently been proposed as a drop‐in replacement for the IEEE‐754 floating‐point representation. We survey the state‐of‐the‐art and state‐of‐the‐practice in the development and use of posits in edge computing and HPC. The current literature supports posits as a promising alternative to traditional floating‐point systems, both as a stand‐alone replacement and in a mixed‐precision environment. Development and standardization of the posit type is ongoing, and much research remains to explore the application of posits in different domains, how to best implement them in hardware, and where they fit with other numerical representations.

Photoactive Electro‐Controlled Visual Perception Memory for Emulating Synaptic Metaplasticity and Hebbian Learning

AbstractIn bionic technology, it has become an innovative process imitating the functionality and structuralism of human biological systems to exploit advanced artificial intelligent machines. Bionics plays a significant role in environmental protection, especially for its low energy loss. By fusing the concept of receptor‐like sensing component and synapse‐like memory, the photoactive electro‐controlled optical sensory memory (PE‐SM) is proposed and realized in a single device, which endows a simple methodology of reducing power consumption by photoactive electro‐control. The PE‐SM is the system built with the stacked atomically thick materials, in which rhenium diselenide serves as a robust photosensor, hexagonal boron nitride serves as a tunneling dielectric, and graphene serves as a charge‐storage layer. With the features of the PE‐SM, it performs synaptic metaplasticities under optical spikes. In addition, a simulated spiking neural network composed of 24 × 24 PE‐SMs is further presented in an unsupervised machine learning environment, performing image recognition via the Hebbian rule. The PE‐SM not only improves the neuromorphic computing efficiency but also simplifies the circuit‐size structure. Eventually, the concept of photoactive electro‐control can extend to other photosensitive 2D materials and provide a new approach of constructing either visual perception memory or photonic synaptic devices.

Retraction: Identifying non-contact defects in fault bearing (IOP Conf. Ser.: Mater. Sci. Eng. 1166 012059)

This article has been retracted by IOP Publishing following an allegation that this article may contain tortured phrases [1], masking overlap of other work [2,3]. IOP Publishing has investigated in line with the COPE guidelines, and agree this article should be retracted. Despite initially responding to our queries, IOP Publishing Limited have not received a response from the authors regarding this retraction. The authors are encouraged to contact IOP Publishing Limited if they wish to contest this retraction. IOP Publishing wishes to credit PubPeer commenters [4] for bringing the issue to our attention. [1] Cabanac G, Labbe C, Magazinov A, 2021, arXiv:2107.06751v1 [2] Goyal D, Dhami S.S, Pabla B.S, 2020, Non-Contact Fault Diagnosis of Bearings in Machine Learning Environment, IEEE Sensors Journal, 20. 9. [3] Goyal D, Choudhary A, Pabla B.S, Dhami S.S (2020) Support vector machines based non-contact fault diagnosis system for bearings, Journal of Intelligent Manufacturing. 31. 1275-1289 [4] https://pubpeer.com/publications/A1DD4EF1B53C429D1165EB0594D102 Retraction published: 06 May 2022

Classification of Autism Spectrum Disorder From EEG-Based Functional Brain Connectivity Analysis.

Autism is a psychiatric condition that is typically diagnosed with behavioral assessment methods. Recent years have seen a rise in the number of children with autism. Since this could have serious health and socioeconomic consequences, it is imperative to investigate how to develop strategies for an early diagnosis that might pave the way to an adequate intervention. In this study, the phase-based functional brain connectivity derived from electroencephalogram (EEG) in a machine learning framework was used to classify the children with autism and typical children in an experimentally obtained data set of 12 autism spectrum disorder (ASD) and 12 typical children. Specifically, the functional brain connectivity networks have quantitatively been characterized by graph-theoretic parameters computed from three proposed approaches based on a standard phase-locking value, which were used as the features in a machine learning environment. Our study was successfully classified between two groups with approximately 95.8% accuracy, 100% sensitivity, and 92% specificity through the trial-averaged phase-locking value (PLV) approach and cubic support vector machine (SVM). This work has also shown that significant changes in functional brain connectivity in ASD children have been revealed at theta band using the aggregated graph-theoretic features. Therefore, the findings from this study offer insight into the potential use of functional brain connectivity as a tool for classifying ASD children.

AI Driven Planetary Bot for Future Vision

These days focus is more on technologies like Artificial Intelligence, Machine Learning and IoT. There is lots of platforms available for IOT implementation. ESP8266 chip is among them Here the implementation is about prediction of different aspects of weather data that can be used in many ways like predicting the future condition of different region of earth or predicting future condition of different planets and their different regions. To implement this system, we need different sensors like pressure sensor humidity sensor, temperature sensor and a light intensity sensor i.e DHT11 is utilize for temperature and humidity data together and LDR. Is for light intensity. The data which is sensed by different sensors are than uploaded to Thingspeak which is an API for cloud server by the help of NodeMCU and then converted to csv format. The data can be used for monitoring the real time values too. Machine Learning Environment can be setup by the help of a CNN model. Training of model can be done by recorded values of sensor data. After recording data from sensors to NodeMCU like temperature, pressure, humidity and light intensity and after these values are sent to python environment that is Jupyter notebook for further analysis. Here the data which is used is real time data to predict the particular value and test the model.

SmartPARE: An R Package for Efficient Identification of True mRNA Cleavage Sites.

Degradome sequencing is commonly used to generate high-throughput information on mRNA cleavage sites mediated by small RNAs (sRNA). In our datasets of potato (Solanum tuberosum, St) and Phytophthora infestans (Pi), initial predictions generated high numbers of cleavage site predictions, which highlighted the need of improved analytic tools. Here, we present an R package based on a deep learning convolutional neural network (CNN) in a machine learning environment to optimize discrimination of false from true cleavage sites. When applying smartPARE to our datasets on potato during the infection process by the late blight pathogen, 7.3% of all cleavage windows represented true cleavages distributed on 214 sites in P. infestans and 444 sites in potato. The sRNA landscape of the two organisms is complex with uneven sRNA production and cleavage regions widespread in the two genomes. Multiple targets and several cases of complex regulatory cascades, particularly in potato, was revealed. We conclude that our new analytic approach is useful for anyone working on complex biological systems and with the interest of identifying cleavage sites particularly inferred by sRNA classes beyond miRNAs.

Detection of Plastic Greenhouses Using High Resolution Rgb Remote Sensing Data and Convolutional Neural Network

Abstract Agricultural production in greenhouses shows a rapid growth in many parts of the world. This form of intensive farming requires a large amount of water and fertilizers, and can have a severe impact on the environment. The number of greenhouses and their location is important for applications like spatial planning, environmental protection, agricultural statistics and taxation. Therefore, with this study we aim to develop a methodology to detect plastic greenhouses in remote sensing data using machine learning algorithms. This research presents the results of the use of a convolutional neural network for automatic object detection of plastic greenhouses in high resolution remotely sensed data within a GIS environment with a graphical interface to advanced algorithms. The convolutional neural network is trained with manually digitized greenhouses and RGB images downloaded from Google Earth. The ArcGIS Pro geographic information system provides access to many of the most advanced python-based machine learning environments like Keras – TensorFlow, PyTorch, fastai and Scikit-learn. These libraries can be accessed via a graphical interface within the GIS environment. Our research evaluated the results of training and inference of three different convolutional neural networks. Experiments were executed with many settings for the backbone models and hyperparameters. The performance of the three models in terms of detection accuracy and time required for training was compared. The model based on the VGG_11 backbone model (with dropout) resulted in an average accuracy of 79.2% with a relatively short training time of 90 minutes, the much more complex DenseNet121 model was trained in 16.5 hours and showed a result of 79.1%, while the ResNet18 based model showed an average accuracy of 83.1% with a training time of 3.5 hours.