Tid: 23 februari 2009 kl 15.15-16.00
Plats : Seminarierummet 3733, Institutionen för matematik, KTH, Lindstedts väg 25, plan 7. Karta!
Föredragshållare: Akoo Hematbolland
Titel: Resource optimization in embedded systems based on data mining. (Examensarbete - Master thesis)
Sammanfattning: This master thesis discusses the resource optimization in embedded systems for Scania's trucks. It is about analyzing historical sale data to find out more information about customers choice. Big part of the work has focus upon studying appropriate methods and tools to the analysis. Since we are dealing with large amount of data, data mining techniques has been used to find relevant information about customer choice. Methods and data mining tools (loglikelihood distance and two step clustering) have been tested on five different functions. With Data Mining a company's knowledge about customer's choice can reduce costs and improve the value of customer relationships.
|Sidansvarig: Filip Lindskog