scholarly journals Smart Use Case Picking with DUCAR: A Hands-On Approach for a Successful Integration of Machine Learning in Production Processes

2020 ◽  
Vol 51 ◽  
pp. 1311-1318
Author(s):  
Franziska Schäfer ◽  
Andreas Mayr ◽  
Erik Schwulera ◽  
Jörg Franke
Keyword(s):  
Machine Learning ◽  
Use Case ◽  
Production Processes ◽  
Hands On ◽  
Successful Integration

scholarly journals Platform Patterns—Using Proven Principles to Develop Digital Platforms

2021 ◽  
Author(s):  
Marvin Drewel ◽  
Leon Özcan ◽  
Jürgen Gausemeier ◽  
Roman Dumitrescu
Keyword(s):  
Mechanical Engineering ◽  
Early Stage ◽  
Use Case ◽  
Digital Platforms ◽  
Hands On ◽  
The Future ◽  
Platform Development ◽  
Strategic Options ◽  
Generic Design ◽  
Platform Economy

AbstractHardly any other area has as much disruptive potential as digital platforms in the course of digitalization. After serious changes have already taken place in the B2C sector with platforms such as Amazon and Airbnb, the B2B sector is on the threshold to the so-called platform economy. In mechanical engineering, pioneers like GE (PREDIX) and Claas (365FarmNet) are trying to get their hands on the act. This is hardly a promising option for small and medium-sized companies, as only a few large companies will survive. Small and medium-sized enterprises (SMEs) are already facing the threat of losing direct consumer contact and becoming exchangeable executers. In order to prevent this, it is important to anticipate at an early stage which strategic options exist for the future platform economy and which adjustments to the product program should already be initiated today. Basically, medium-sized companies in particular lack a strategy for an advantageous entry into the future platform economy.The paper presents different approaches to master the challenges of participating in the platform economy by using platform patterns. Platform patterns represent proven principles of already existing platforms. We show how we derived a catalogue with 37 identified platform patterns. The catalogue has a generic design and can be customized for a specific use case. The versatility of the catalogue is underlined by three possible applications: (1) platform ideation, (2) platform development, and (3) platform characterization.

scholarly journals Construction of a quality model for machine learning systems

2021 ◽  
Author(s):  
Julien Siebert ◽  
Lisa Joeckel ◽  
Jens Heidrich ◽  
Adam Trendowicz ◽  
Koji Nakamichi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Lessons Learned ◽  
Training Data ◽  
Use Case ◽  
Construction Process ◽  
Quality Model ◽  
Quality Models ◽  
Quality Properties ◽  
Reference Quality ◽  
Industrial Use

AbstractNowadays, systems containing components based on machine learning (ML) methods are becoming more widespread. In order to ensure the intended behavior of a software system, there are standards that define necessary qualities of the system and its components (such as ISO/IEC 25010). Due to the different nature of ML, we have to re-interpret existing qualities for ML systems or add new ones (such as trustworthiness). We have to be very precise about which quality property is relevant for which entity of interest (such as completeness of training data or correctness of trained model), and how to objectively evaluate adherence to quality requirements. In this article, we present how to systematically construct quality models for ML systems based on an industrial use case. This quality model enables practitioners to specify and assess qualities for ML systems objectively. In addition to the overall construction process described, the main outcomes include a meta-model for specifying quality models for ML systems, reference elements regarding relevant views, entities, quality properties, and measures for ML systems based on existing research, an example instantiation of a quality model for a concrete industrial use case, and lessons learned from applying the construction process. We found that it is crucial to follow a systematic process in order to come up with measurable quality properties that can be evaluated in practice. In the future, we want to learn how the term quality differs between different types of ML systems and come up with reference quality models for evaluating qualities of ML systems.

Meta-analysis and machine learning-augmented mixed effects cohort analysis of improved diets among 5,847 medical trainees, providers, and patients

2021 ◽  
pp. 1-27
Author(s):  
Dominique J. Monlezun ◽  
Christopher Carr ◽  
Tianhua Niu ◽  
Francesco Nordio ◽  
Nicole DeValle ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cohort Study ◽  
Nutrition Education ◽  
Meta Analysis ◽  
Cohort Analysis ◽  
Cost Effective ◽  
Number Needed To Treat ◽  
Nutrition Counseling ◽  
Medical Trainees ◽  
Hands On

Abstract Objective: We sought to produce the first meta-analysis (of medical trainee competency improvement in nutrition counseling) informing the first cohort study of patient diet improvement through medical trainees and providers counseling patients on nutrition. Design: (Part A) A systematic review and meta-analysis informing (Part B) the intervention analyzed in the world’s largest prospective multi-center cohort study on hands-on cooking and nutrition education for medical trainees, providers, and patients. Settings: (A) Medical educational institutions. (B) Teaching kitchens. Participants: (A) Medical trainees. (B) Trainees, providers, and patients. Results: (A) Of the 212 citations identified (N=1,698 trainees), 11 studies met inclusion criteria. The overall effect size was 9.80 (95%CI 7.15-12.456.87-13.85; p<0.001), comparable to the machine learning (ML)-augmented results. The number needed to treat for the top performing high quality study was 12. (B) The hands-on cooking and nutrition education curriculum from the top performing study was applied for medical trainees and providers who subsequently taught patients in the same curriculum (N=5,847). The intervention compared to standard medical care and education alone significantly increased the odds of superior diets (high/medium versus low Mediterranean diet adherence) for residents/fellows most (OR 10.79, 95%CI 4.94-23.58; p<0.001) followed by students (OR 9.62, 95%CI 5.92-15.63; p<0.001), providers (OR 5.19, 95%CI 3.23-8.32, p<0.001), and patients (OR 2.48, 95%CI 1.38-4.45; p=0.002), results consistent with those from ML. Conclusions: This study suggests that medical trainees and providers can improve patients’ diets with nutrition counseling in a manner that is clinically and cost effective and may simultaneously advance societal equity.

Hands-on with Azure Machine Learning

2020 ◽  
pp. 149-199
Author(s):  
Julian Soh ◽  
Priyanshi Singh
Keyword(s):  
Machine Learning ◽  
Hands On

scholarly journals Phybrata Sensors and Machine Learning for Enhanced Neurophysiological Diagnosis and Treatment

Sensors ◽  
2021 ◽  
Vol 21 (21) ◽  
pp. 7417
Author(s):  
Alex J. Hope ◽  
Utkarsh Vashisth ◽  
Matthew J. Parker ◽  
Andreas B. Ralston ◽  
Joshua M. Roper ◽  
...  
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Random Forest ◽  
Binary Classification ◽  
Classification Performance ◽  
Support Vector ◽  
Use Case ◽  
Signal Features ◽  
Test Population

Concussion injuries remain a significant public health challenge. A significant unmet clinical need remains for tools that allow related physiological impairments and longer-term health risks to be identified earlier, better quantified, and more easily monitored over time. We address this challenge by combining a head-mounted wearable inertial motion unit (IMU)-based physiological vibration acceleration (“phybrata”) sensor and several candidate machine learning (ML) models. The performance of this solution is assessed for both binary classification of concussion patients and multiclass predictions of specific concussion-related neurophysiological impairments. Results are compared with previously reported approaches to ML-based concussion diagnostics. Using phybrata data from a previously reported concussion study population, four different machine learning models (Support Vector Machine, Random Forest Classifier, Extreme Gradient Boost, and Convolutional Neural Network) are first investigated for binary classification of the test population as healthy vs. concussion (Use Case 1). Results are compared for two different data preprocessing pipelines, Time-Series Averaging (TSA) and Non-Time-Series Feature Extraction (NTS). Next, the three best-performing NTS models are compared in terms of their multiclass prediction performance for specific concussion-related impairments: vestibular, neurological, both (Use Case 2). For Use Case 1, the NTS model approach outperformed the TSA approach, with the two best algorithms achieving an F1 score of 0.94. For Use Case 2, the NTS Random Forest model achieved the best performance in the testing set, with an F1 score of 0.90, and identified a wider range of relevant phybrata signal features that contributed to impairment classification compared with manual feature inspection and statistical data analysis. The overall classification performance achieved in the present work exceeds previously reported approaches to ML-based concussion diagnostics using other data sources and ML models. This study also demonstrates the first combination of a wearable IMU-based sensor and ML model that enables both binary classification of concussion patients and multiclass predictions of specific concussion-related neurophysiological impairments.

Audience Meter: a Use Case of Deploying Machine Learning Algorithms over 5G Networks with MEC

2020 ◽  
Author(s):  
Ester Gonzalez-Sosa ◽  
Ignacio Frontelo-Benito ◽  
Redouane Kachach ◽  
Pablo Perez ◽  
Jaime Jesus Ruiz ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Use Case ◽  
5G Networks
Sign in / Sign up

Export Citation Format

Share Document