Myopic Policies in Sequential Classification

Moshe Ben-Bassat

doi:10.1109/TC.1978.1675054

Myopic Policies in Sequential Classification

Moshe Ben-Bassat^*

^*Corresponding author for this work

Coller School of Management

Research output: Contribution to journal › Article › peer-review

30 Scopus citations

Abstract

Several rules for feature selection in myopic policy are examined for solving the sequential finite classification problem with conditionally independent binary features. The main finding is that no rule is consistently superior to the others. Likewise no specific strategy for the alternating of rules seems to be significantly more efficient.

Original language	English
Pages (from-to)	170-174
Number of pages	5
Journal	IEEE Transactions on Computers
Volume	C-27
Issue number	2
DOIs	https://doi.org/10.1109/TC.1978.1675054
State	Published - Feb 1978

Access to Document

10.1109/TC.1978.1675054

Cite this

@article{5e9e9ca2d2a9447a9ef6a91fba3dbaa0,

title = "Myopic Policies in Sequential Classification",

abstract = "Several rules for feature selection in myopic policy are examined for solving the sequential finite classification problem with conditionally independent binary features. The main finding is that no rule is consistently superior to the others. Likewise no specific strategy for the alternating of rules seems to be significantly more efficient.",

author = "Moshe Ben-Bassat",

year = "1978",

month = feb,

doi = "10.1109/TC.1978.1675054",

language = "אנגלית",

volume = "C-27",

pages = "170--174",

journal = "IEEE Transactions on Computers",

issn = "0018-9340",

publisher = "IEEE Computer Society",

number = "2",

}

TY - JOUR

T1 - Myopic Policies in Sequential Classification

AU - Ben-Bassat, Moshe

PY - 1978/2

Y1 - 1978/2

N2 - Several rules for feature selection in myopic policy are examined for solving the sequential finite classification problem with conditionally independent binary features. The main finding is that no rule is consistently superior to the others. Likewise no specific strategy for the alternating of rules seems to be significantly more efficient.

AB - Several rules for feature selection in myopic policy are examined for solving the sequential finite classification problem with conditionally independent binary features. The main finding is that no rule is consistently superior to the others. Likewise no specific strategy for the alternating of rules seems to be significantly more efficient.

UR - http://www.scopus.com/inward/record.url?scp=0017933553&partnerID=8YFLogxK

U2 - 10.1109/TC.1978.1675054

DO - 10.1109/TC.1978.1675054

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0017933553

SN - 0018-9340

VL - C-27

SP - 170

EP - 174

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

IS - 2

ER -

Myopic Policies in Sequential Classification

Abstract

Access to Document

Other files and links

Fingerprint

Cite this