Prediction of Successful Participation in a Lifestyle Activity Program using Data Mining Techniques

Created by W.Langdon from gp-bibliography.bib Revision:1.7615

@InProceedings{Pijl:2009:BNAIC,

• author = "Marten Pijl and Joyca Lacroix and Steffen Pauws and Annelies Goris",
• title = "Prediction of Successful Participation in a Lifestyle Activity Program using Data Mining Techniques",
• booktitle = "The 21st Benelux Conference on Artificial Intelligence (BNAIC)",
• year = "2009",
• editor = "Toon Calders and Karl Tuyls and Mykola Pechenizkiy",
• address = "Eindhoven",
• month = "29-30 " # oct,
• organisation = "BNVKI",
• note = "Industry Track",
• keywords = "genetic algorithms, genetic programming",
• annote = "The Pennsylvania State University CiteSeerX Archives",
• bibsource = "OAI-PMH server at citeseerx.ist.psu.edu",
• language = "en",
• oai = "oai:CiteSeerX.psu:10.1.1.380.4042",
• rights = "Metadata may be used without restrictions as long as the oai identifier remains attached to it.",
• URL = "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.380.4042",
• URL = "http://wwwis.win.tue.nl/bnaic2009/papers/bnaic2009_paper_114.pdf",
• size = "8 pages",
• abstract = "The growing number of people worldwide living a sedentary life has led to increased efforts into the development of effective physical activity intervention programs. However, many participants of such programs fail to complete the program, and as a result do not attain the desired increase in physical activity. By detecting participants who are at risk of dropping out in advance, it may be possible to intervene and prevent the participant from dropping out. The work in this paper discusses the classification of such participants, using an activity database containing participant characteristics and activity data gathered through an accelerometer worn by the participants. Using genetic programming techniques, database variables (called markers) are combined to form new sets of markers. The predictive power of these new markers are compared to the markers present in the database, based on classification using principal component analysis and a k-nearest neighbour classifier. Results show that without the use of genetic programming to combine markers, classification class accuracy is approximately 64percent. By combining markers through genetic programming, classification class accuracy increases to 72percent, which equates to a reduction in the number of errors of approximately 23percent",
• notes = "Philips",

}

Genetic Programming entries for Marten Pijl Joyca Lacroix Steffen Pauws Annelies Goris

Citations