Prediction strategies: Combining prediction techniques to optimize personalization

Mark van Setten; L. Ardissono; A. Buczak; Mettina Veenstra; Antinus Nijholt

Prediction strategies: Combining prediction techniques to optimize personalization

Mark van Setten, L. Ardissono (Editor), A. Buczak (Editor), Mettina Veenstra, Antinus Nijholt

Research output: Contribution to conference › Paper › peer-review

1 Downloads (Pure)

Abstract

An important step in providing personalized information, such as personalized electronic program guides, is predicting the level of interest for a specific user in information, such as TV programs. This paper describes a model that can be used to combine prediction techniques, which predict the level of interest, into prediction strategies. Our hypothesis is that prediction strategies optimize the prediction quality and that they lead to more reliable predictions, because they take into account the state of the system, the users and the information. Results of an initial experiment support this hypothesis.

Original language	Undefined
Pages	29-37
Number of pages	9
Publication status	Published - 12 May 2002
Event	2nd Workshop on Personalisation in Future TV, TV-02 - Universidad de Málaga, Málaga, Spain Duration: 12 May 2002 → 12 May 2002 Conference number: 2

Workshop

Workshop	2nd Workshop on Personalisation in Future TV, TV-02
Abbreviated title	TV-02
Country/Territory	Spain
City	Málaga
Period	12/05/02 → 12/05/02

Keywords

HMI-IE: Information Engineering
EWI-6675
IR-63370

Access to Document

http://eprints.eemcs.utwente.nl/secure2/6675/01/umtv2002.pdf

Cite this

@conference{e2e36a8fc849450ebe2a37182dc20f5b,

title = "Prediction strategies: Combining prediction techniques to optimize personalization",

abstract = "An important step in providing personalized information, such as personalized electronic program guides, is predicting the level of interest for a specific user in information, such as TV programs. This paper describes a model that can be used to combine prediction techniques, which predict the level of interest, into prediction strategies. Our hypothesis is that prediction strategies optimize the prediction quality and that they lead to more reliable predictions, because they take into account the state of the system, the users and the information. Results of an initial experiment support this hypothesis.",

keywords = "HMI-IE: Information Engineering, EWI-6675, IR-63370",

author = "{van Setten}, Mark and L. Ardissono and A. Buczak and Mettina Veenstra and Antinus Nijholt",

year = "2002",

month = may,

day = "12",

language = "Undefined",

pages = "29--37",

note = "2nd Workshop on Personalisation in Future TV, TV-02 ; Conference date: 12-05-2002 Through 12-05-2002",

}

TY - CONF

T1 - Prediction strategies: Combining prediction techniques to optimize personalization

AU - van Setten, Mark

AU - Veenstra, Mettina

AU - Nijholt, Antinus

A2 - Ardissono, L.

A2 - Buczak, A.

N1 - Conference code: 2

PY - 2002/5/12

Y1 - 2002/5/12

N2 - An important step in providing personalized information, such as personalized electronic program guides, is predicting the level of interest for a specific user in information, such as TV programs. This paper describes a model that can be used to combine prediction techniques, which predict the level of interest, into prediction strategies. Our hypothesis is that prediction strategies optimize the prediction quality and that they lead to more reliable predictions, because they take into account the state of the system, the users and the information. Results of an initial experiment support this hypothesis.

AB - An important step in providing personalized information, such as personalized electronic program guides, is predicting the level of interest for a specific user in information, such as TV programs. This paper describes a model that can be used to combine prediction techniques, which predict the level of interest, into prediction strategies. Our hypothesis is that prediction strategies optimize the prediction quality and that they lead to more reliable predictions, because they take into account the state of the system, the users and the information. Results of an initial experiment support this hypothesis.

KW - HMI-IE: Information Engineering

KW - EWI-6675

KW - IR-63370

M3 - Paper

SP - 29

EP - 37

T2 - 2nd Workshop on Personalisation in Future TV, TV-02

Y2 - 12 May 2002 through 12 May 2002

ER -