scholarly journals Design Thinking and Knowledge Engineering: A Machine Learning Case

2020 ◽  
Vol 10 (6) ◽  
pp. 765-770
Author(s):  
Michael Walch ◽  
◽  
Dimitris Karagiannis
1992 ◽  
Author(s):  
Christopher A. Miller ◽  
Keith R. Levi ◽  
Barry Druhan ◽  
Valerie L. Shalin

Author(s):  
Devdas Shetty ◽  
Jiajun Xu

It is suggested by many scholars that if the goal of engineering education is to produce engineers who can critically design and create, then providing students with early opportunities to engage in creative engineering design is important. While basic design is focused on the development of new products for the individual, working towards a more sustainable world demands greater attention to designing for and with communities. Improving design education and examining design-learning outcomes requires a kind of targeted approach that could match the best practices to personalize student learning. Design is complex and design includes balancing the needs of multiple stakeholders. However, there is a gap in the preparation of design education that will be needed in a challenging environment. This paper reviews the history of design thinking in the engineering curriculum. Design thinking education starts with an understanding of its importance with socioeconomic relevance. Through observation and empathy, mapping the designer uses the listening and learning tools for mapping users unarticulated needs, working in a team environment. The designer takes time to think carefully why a certain project is considered and details which aspects of machine learning application can be applied from functional to complete success for the end users. The availability of powerful virtual reality methodologies, have made it possible to consider the realistic needs and visualize scenarios and to explore the design alternatives with new ideas before full scale resource allocation on new ideas. Mid-to-advanced level courses with experimental assignments require that students apply through experimentation the principles and concepts learned in foundation courses. The basic design tools such as axiomatic thinking, theory of inventive problem solving, design iteration and simulation using hardware-in-the loop are discussed with case studies. Consideration of product sustainability with the thoughts of design for disassembly and disposal has emerged as a major part of design thinking. Senior engineering courses center on cross and interdisciplinary design and capstone experiences so that students experience fully guided practice of device design and problem solving, simulating what they are likely to experience in the world. This paper examines the critical issues of design thinking in a curriculum from observation, empathy mapping, validation of the idea, and improvement of idea by virtual reality and machine learning, optimization of the idea by tools such as axiomatic design, hardware in the loop simulation, and finally examining product sustainability causes.


Author(s):  
Nattaphol Thanachawengsakul ◽  
Panita Wannapiroon ◽  
Prachyanun Nilsook

The knowledge repository management system architecture of digital knowledge engineering using machine learning (KRMS-SWE) to promote software engineering competencies is comprised of four parts, as follows: 1) device service, 2) application service, 3) module service of the KRMS-SWE and 4) machine learning service and storage unit. The knowledge creation, storage, testing and assessing of students’ knowledge in software engineering is carried out using a knowledge verification process with machine learning and divided into six steps, as follows: pre-processing, filtration, stemming, indexing, data mining and interpretation and evaluation. The overall result regarding the suitability of the KRMS-SWE is assessed by five experts who have high levels of experience in related fields. The findings reveal that this research approach can be applied to the future development of the KRMS-SWE.


Data Mining ◽  
2013 ◽  
pp. 1979-1996
Author(s):  
Klaus Wölfel ◽  
Jean-Paul Smets

Free/Open Source software (FOSS) has made Enterprise Resource Planning (ERP) systems more accessible for Small and Medium Enterprises (SMEs) including overseas subsidiaries of large companies. However, the consulting required to configure an ERP to meet the specific needs of an organization remains a major financial and organizational burden for SMEs. Automatic ERP package configuration based on knowledge engineering, machine learning and data mining could be a solution to lessen the burden of the implementation process. This chapter presents two approaches to an automation of selected configuration options of the FOS-ERP package ERP5. These approaches are based on knowledge engineering with decision trees and machine learning with classifiers. The design of the ERP5 Artificial intelligence Toolkit (EAT) aims at the integration of these approaches into ERP5. The chapter also shows how FOS-ERP can boost Information System (IS) research. The investigation of the automation approaches was only possible because the free source code and technical documentation of ERP5 was accessible for TU Dresden researchers.


Sign in / Sign up

Export Citation Format

Share Document