Machine Learning Algorithms for Recommending Design Methods

2014 ◽  
Vol 136 (10) ◽  
Author(s):  
Mark Fuge ◽  
Bud Peters ◽  
Alice Agogino
Keyword(s):  
Case Studies ◽  
Spectral Clustering ◽  
Design Methods ◽  
Machine Learning Algorithms ◽  
New Methods ◽  
Recommendation Algorithms ◽  
Problem Description ◽  
Novice Designers ◽  
Effective Product

Every year design practitioners and researchers develop new methods for understanding users and solving problems. This increasingly large collection of methods causes a problem for novice designers: How does one choose which design methods to use for a given problem? Experienced designers can provide case studies that document which methods they used, but studying these cases to infer appropriate methods for a novel problem is inefficient. This research addresses that issue by applying techniques from content-based and collaborative filtering to automatically recommend design methods, given a particular problem. Specifically, we demonstrate the quality with which different algorithms recommend 39 design methods out of an 800+ case study dataset. We find that knowing which methods occur frequently together allows one to recommend design methods more effectively than just using the text of the problem description itself. Furthermore, we demonstrate that automatically grouping frequently co-occurring methods using spectral clustering replicates human-provided groupings to 92% accuracy. By leveraging existing case studies, recommendation algorithms can help novice designers efficiently navigate the increasing array of design methods, leading to more effective product design.

Patientenbezogene Komplexität in der Pflege – Kollektive Case Studies im Akutspital

Pflege ◽  
2018 ◽  
Vol 31 (5) ◽  
pp. 237-244 ◽  
Author(s):  
Caroline Gurtner ◽  
Rebecca Spirig ◽  
Diana Staudacher ◽  
Evelyn Huber
Keyword(s):  
Case Studies ◽  
Study Designs

Zusammenfassung. Hintergrund: Die patientenbezogene Komplexität der Pflege ist durch die Merkmale „Instabilität“, „Unsicherheit“ und „Variabilität“ definiert. Aufgrund der reduzierten Aufenthaltsdauer und der steigenden Zahl chronisch und mehrfach erkrankter Personen erhöht sich die Komplexität der Pflege. Ziel: In dieser Studie untersuchten wir das Phänomen patientenbezogener Komplexität aus Sicht von Pflegefachpersonen und Pflegeexpertinnen im Akutspital. Methode: Im Rahmen eines kollektiven Case-Study-Designs schätzten Pflegefachpersonen und Pflegeexpertinnen die Komplexität von Pflegesituationen mit einem Fragebogen ein. Danach befragten wir sie in Einzelinterviews zu ihrer Einschätzung. Mittels Within-Case-Analyse verdichteten wir die Daten induktiv zu Fallgeschichten. In der Cross-Case-Analyse verglichen wir die Fallgeschichten hinsichtlich deduktiv abgeleiteter Merkmale. Ergebnisse: Die Ausprägung der Komplexität hing in den vier Cases im Wesentlichen davon ab, ob klinische Probleme kontrollierbar und prognostizierbar waren. Je nach individuellen Ressourcen der Patientinnen und Patienten stieg bzw. sank die Komplexität. Schlussfolgerungen: Komplexe Patientensituationen fordern von Pflegefachpersonen Fachwissen, Erfahrung, kommunikative Kompetenzen sowie die Fähigkeit zur Reflexion. Berufsanfänger und Berufsanfängerinnen werden zur Entwicklung dieser Fähigkeiten idealerweise durch erfahrene Berufskolleginnen oder -kollegen unterstützt und beraten.

SenCity – Evaluating Users’ Experiences of Intelligent Lighting for Well-Being in Smart Cities

2018 ◽  
pp. 60-67
Author(s):  
Henrika Pihlajaniemi ◽  
Anna Luusua ◽  
Eveliina Juntunen
Keyword(s):  
Case Studies ◽  
Smart City ◽  
Smart Cities ◽  
Well Being ◽  
Evaluation Methods ◽  
Data Analyses ◽  
Sense Of Security ◽  
System Data

This paper presents the evaluation of usersХ experiences in three intelligent lighting pilots in Finland. Two of the case studies are related to the use of intelligent lighting in different kinds of traffic areas, having emphasis on aspects of visibility, traffic and movement safety, and sense of security. The last case study presents a more complex view to the experience of intelligent lighting in smart city contexts. The evaluation methods, tailored to each pilot context, include questionnaires, an urban dashboard, in-situ interviews and observations, evaluation probes, and system data analyses. The applicability of the selected and tested methods is discussed reflecting the process and achieved results.

Nonwovens: Global Trends in World Economy, European Foreign Trade, and Selected Case Studies from Poland and Asian Brics Countries

2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Zofia Wysokińska ◽  
Tomasz Czajkowski ◽  
Katarzyna Grabowska
Keyword(s):  
Case Studies ◽  
World Economy ◽  
World Market ◽  
Global Trends ◽  
Textile Materials ◽  
Brics Countries ◽  
Recent Trends ◽  
Almost All ◽  
Special Case

AbstractNonwovens are one of the most versatile textile materials and have become increasingly popular in almost all sectors of the economy due to their low manufacturing costs and unique properties. In the next few years, the world market of nonwovens is predicted to grow by 7%–8% annually (International Nonwovens & Disposables Association [INDA], European Disposables and Nonwovens Association [EDANA], and Markets and Markets). This article aims to analyze the most recent trends in the global export and import of nonwovens, to present two case studies of Polish companies that produce them, and to present one special case study of the market of nonwoven geotextiles in China and India, which are the Asian transition economies among the BRICS countries (Brazil, Russia, India, China, and South Africa).

Deep Reinforcement Learning for Multiparameter Optimization in de novo Drug Design

2019 ◽  
Author(s):  
Niclas Ståhl ◽  
Göran Falkman ◽  
Alexander Karlsson ◽  
Gunnar Mathiason ◽  
Jonas Boström
Keyword(s):  
Reinforcement Learning ◽  
Short Term Memory ◽  
De Novo ◽  
De Novo Drug Design ◽  
Generative Process ◽  
New Methods ◽  
Multiparameter Optimization ◽  
Long Short Term Memory ◽  
New Compounds

<p>In medicinal chemistry programs it is key to design and make compounds that are efficacious and safe. This is a long, complex and difficult multi-parameter optimization process, often including several properties with orthogonal trends. New methods for the automated design of compounds against profiles of multiple properties are thus of great value. Here we present a fragment-based reinforcement learning approach based on an actor-critic model, for the generation of novel molecules with optimal properties. The actor and the critic are both modelled with bidirectional long short-term memory (LSTM) networks. The AI method learns how to generate new compounds with desired properties by starting from an initial set of lead molecules and then improve these by replacing some of their fragments. A balanced binary tree based on the similarity of fragments is used in the generative process to bias the output towards structurally similar molecules. The method is demonstrated by a case study showing that 93% of the generated molecules are chemically valid, and a third satisfy the targeted objectives, while there were none in the initial set.</p>

Application of KOH Electrochemical Etch and Passive Voltage Contrast Techniques for Deep Sub-Micron CMOS Leaky Gate Detection

2000 ◽  
Author(s):  
Fred Y. Chang ◽  
Victer Chan
Keyword(s):  
Case Studies ◽  
Detection Technique ◽  
Cobalt Silicide ◽  
Process Technology ◽  
Process Flow ◽  
Voltage Contrast ◽  
Emission Detection ◽  
Wet Etch ◽  
Direct Confirmation

Abstract This paper describes a novel de-process flow by combining cobalt silicide / nitride wet etch with KOH electrochemical wet etch (ECW) to identify leaky gate in silicided deep sub-micron process technology. Traditionally, leaky gate identification requires direct confirmation by gate level electrical or emission detection technique. Ohtani [1] used KOH electrochemical etch application to identify nonsilicided leaky gate capacitor in DRAM without using the above confirmation. The result of the case study demonstrates the expanded application of ECW etch to both silicided 0.18um logic and SRAM devices. Voltage contrast at metal 1 to assist leaky gate localization is also proposed. By combining both techniques, the possibility for isolating gate related defects are greatly enhanced. Case studies also show the advantages of the proposed technique over conventional poly level voltage contrast in leaky gate identification especially with devices that use local interconnect and nitride liner process.

Sign in / Sign up

Export Citation Format

Share Document