GLOSSARY ENTRY (DERIVED FROM QUESTION BELOW) | ||||||
---|---|---|---|---|---|---|
|
20:13 Sep 24, 2004 |
German to English translations [PRO] Social Sciences - Mathematics & Statistics | |||||||
---|---|---|---|---|---|---|---|
|
| ||||||
| Selected response from: Brandis (X) Local time: 01:53 | ||||||
Grading comment
|
Summary of answers provided | ||||
---|---|---|---|---|
3 +1 | criteria for differentiation |
| ||
3 | segmentation criterium / splitting criterium |
|
criteria for differentiation Explanation: criteria for differentiation - I think, that is what we had used in statistics. |
| |
Grading comment
| ||
Login to enter a peer comment (or grade) |
segmentation criterium / splitting criterium Explanation: The first term seems to be used more in a psychological/sociological environment: Segmentation criteria - demographic Age Gender Race and ethnic group Hobbies Lifestyle Reading, listening, viewing patterns Education Social class . http://www.psychologie.uni-mannheim.de/psycho2/proj/wmc/web/... dynaTrack Methodology: Patient Classification and Property Creation Dynamic Patient Segmentation: Effective analytics requires the ability to segment the patient population into meaningful groups and perform in-depth analysis on each. An unrestricted number and types of ‘Properties’ are used to classify masked patient records into multiple Categories. Examples of ‘Properties’ are: * Age, categorizes patients by age range, each category may span 10 years (category ranges can be customized to specific requirements) * Gender, categories by Male, Female and unknown [...] http://www.interconus.com/products_dynatrack_sales.htm Studies of exercise-related attitudes among African American women, for example, point to the importance of audience segmentation. African American women have a greater tolerance for higher body weights than women of European descent [...] They segment the audience into sub-audiences, and tailor messages to each of these groups. Stage of change, ethnic/cultural background, socio-economic status, language, and gender are among the variables that may be appropriate segmentation criteria. http://www.lin.ca/resource/html/interven.htm ========== There's also a statistical tree modell (predictions, decision making) where something similar is used: Q7. Welche Kriterien fuer das Wachstum von Baeumen (splitting criteria) bietet CART? A7. CART beinhaltet sieben Teilungskriterien die nur eine Variable berücksichtigen- Gini, Symgini, Twoing, ordered twoing und "class probability" fuer Klassifikationsbaeume, sowie kleinste Quadrate und kleinste absolute Abweichung fuer Regressions Baeume. Hinzu kommte ein multi-variables Splitting-Kriterium, die 'linear combinations' -Methode. Die voreingestellte Gini-Methode ergibt meist die besten Resultate aber unter bestimmten Umstaenden koennen andere Methoden genauere Modelle liefern. http://www.statcon.de/produkte/CART_5.0.html Families of splitting criteria for classification trees Full text Publisher Site Source Statistics and Computing http://portal.acm.org/citation.cfm?id=599237.599356 =============== |
| |
Login to enter a peer comment (or grade) |
Login or register (free and only takes a few minutes) to participate in this question.
You will also have access to many other tools and opportunities designed for those who have language-related jobs (or are passionate about them). Participation is free and the site has a strict confidentiality policy.