Today's Contents1.KanseiEngineering感性工程.The processes of Kansei Engineering and examples2.MultivariateAnalysis多变量分析: From Kansei evaluation data to models (knowledge)3.Exercise1:CreationofRatingScales评级量表Kanseiwords感性词愁绪MelancholyCheerful快乐1234567Rating scaleBipolarscale双极规模
1. Kansei Engineering 感性工程 • The processes of Kansei Engineering and examples 2. Multivariate Analysis 多变量分析 • From Kansei evaluation data to models (knowledge) 3. Exercise 1: Creation of Rating Scales 评级量表 Today’s Contents Melancholy Cheerful 1 2 3 4 5 6 7 Rating scale 愁绪 快乐 6 Kansei words 感性词 Bipolar scale 双极规模
1. Kansei EngineeringFriendlyA)PreparationofproductsamplesGorgeousB)PreparationofKanseiwordsC) Evaluation experimentsFeminineD) Data analysis:Relationshipsbetweenproduct samplesOld-fashioned:RelationshipsbetweenKanseiwords:RelationshipsbetweenKanseievaluationandproductdesign(Mainpurpose)E)Utilizationofknowledgeobtained:Designingnewproductsanddevelopingsalesstrategies重要！DevelopingrecommendationsystemsforElectronicCommerce
A) Preparation of product samples B) Preparation of Kansei words C) Evaluation experiments D) Data analysis • Relationships between product samples • Relationships between Kansei words • Relationships between Kansei evaluation and product design (Main purpose) E) Utilization of knowledge obtained • Designing new products and developing sales strategies • Developing recommendation systems for Electronic Commerce 7 1. Kansei Engineering Gorgeous Old-fashioned Feminine Friendly
Kansei evaluation and modeling (Overview)WarmSoftKanseiwordsCloudyWhatkindsof wordcanbe usedto:RepresentativeKanseifactorstoevaluatethiscup?evaluatetheproducts·RelationshipsbetweenKansei多变量分析evaluationanddesignelementsMultivariateWhatkindsofanalysisknowledgecanbeProbabilitymodelsofproductsderivedfromthewithrespecttoKanseiwordsProbabilityevaluation data?(Ex.)Theprobabilitythatthismodelingcupisrated"Warm"is70%.概率建模
Kansei evaluation and modeling (Overview) What kinds of word can be used to evaluate this cup? Soft Warm Cloudy What kinds of knowledge can be derived from the evaluation data? Multivariate analysis Probability modeling • Representative Kansei factors to evaluate the products • Relationships between Kansei evaluation and design elements • Probability models of products with respect to Kansei words • (Ex.) The probability that this cup is rated “Warm” is 70%. 8 Kansei words 多变量分析 概率建模
Data collection and data screening·Preparation:Thefirsttaskistoprepareproductsforevaluationbythebipolarscales used in the semantic differential method (Osgood, et al.,1957)..OsgoodCE,SuciGJ,TannenbaumPH(1957)TheMeasurementofMeaning,Universityof IllinoisPress·Experiment:The evaluation experiment should be designed carefully.Dependingontheproducts to be evaluated, wemust collectappropriateevaluators..Notethatthetimeand costtogetgooddataisverylarge.·Screening:Thedata screeningissometimesnecessarybecauseoferrorsorbiased scoring.Moreover,thosewhoarefamiliarwiththeproductsandthosewhodonotknowthemwouldmakedifferentscoresinsomescales.9
Data collection and data screening • Preparation: The first task is to prepare products for evaluation by the bipolar scales used in the semantic differential method (Osgood, et al., 1957). • Osgood CE, Suci GJ, Tannenbaum PH (1957) The Measurement of Meaning, University of Illinois Press. • Experiment: The evaluation experiment should be designed carefully. Depending on the products to be evaluated, we must collect appropriate evaluators. • Note that the time and cost to get good data is very large. • Screening: The data screening is sometimes necessary because of errors or biased scoring. Moreover, those who are familiar with the products and those who do not know them would make different scores in some scales. 9
Bipolar scales1-6CuteTasteful口品口口口QuietBusyOOOOOOOOOODOLightHeavyDelicateOOLODOOSmartStatic000ODynamic口口SemanticDifferentialMethod(Osgood,1957)Kanseievaluationexperiment10
Bipolar scales 1 2 3 4 5 6 7 Cute □■□□□□□ Tasteful Busy □□□□□■□ Quiet Heavy □□□□□□■ Light Delicate □□■□□□□ Smart Static □□□□■□□ Dynamic Semantic Differential Method (Osgood, 1957) Kansei evaluation experiment 10