On improving random forest for hard-to-classify records

Md Nasim Adnan; Md Zahidul Islam

doi:10.1007/978-3-319-49586-6_39

On improving random forest for hard-to-classify records

Md Nasim Adnan, Md Zahidul Islam

Research output: Book chapter/Published conference paper › Conference paper › peer-review

1 Citation (Scopus)

Abstract

Random Forest draws much interest from the research community because of its simplicity and excellent performance. The splitting attribute at each node of a decision tree for Random Forest is determined from a predefined number of randomly selected subset of attributes of the entire attribute set. The size of the subset is one of the most controversial points of Random Forest that encouraged many contributions. However, a little attention is given to improve Random Forest specifically for those records that are hard to classify. In this paper, we propose a novel technique of detecting hard-to-classify records and increase the weights of those records in a training data set. We then build Random Forest from the weighted training data set. The experimental results presented in this paper indicate that the ensemble accuracy of Random Forest can be improved when applied on weighted training data sets with more emphasis on hard-to-classify records.

Original language	English
Title of host publication	Proceedings of the 12th International Conference on Advanced Data Mining and Applications, ADMA 2016
Editors	Jinyan Li, Xue Li, Shuliang Wang, Jianxin Li, Quan Z. Sheng
Place of Publication	Switzerland
Publisher	Springer
Pages	558-566
Number of pages	9
Volume	10086
ISBN (Electronic)	9783319495866
ISBN (Print)	9783319495859
DOIs	https://doi.org/10.1007/978-3-319-49586-6_39
Publication status	Published - 2016
Event	Advanced Data Mining and Applications (ADMA) 12th International Conference - Mantra Legends Hotel, Gold Coast, Australia Duration: 12 Dec 2016 → 15 Dec 2016 https://cs.adelaide.edu.au/~adma2016/ https://cs.adelaide.edu.au/~adma2016/ADMA2016_program.pdf (Conference program)

Conference

Conference	Advanced Data Mining and Applications (ADMA) 12th International Conference
Country/Territory	Australia
City	Gold Coast
Period	12/12/16 → 15/12/16
Other	The year 2016 marks the 12th aniversary of the International Conference on Advanced Data Mining and Applications (ADMA 2016). The conference aims at bringing together the experts on data mining from around the world, and providing a leading international forum for the dissemination of original research findings in data mining, spanning applications, algorithms, software and systems, as well as different applied disciplines with potential in data mining, such as smartphone and social network mining, bio-medical science and green computing. ADMA 2016 will promote the same close interaction and collaboration among practitioners and researchers. Published papers will go through a full peer review process.
Internet address	https://cs.adelaide.edu.au/~adma2016/ https://cs.adelaide.edu.au/~adma2016/ADMA2016_program.pdf (Conference program)

Access to Document

10.1007/978-3-319-49586-6_39

Cite this

@inproceedings{060065c2719a4fb7bb4f71687a8ef7cd,

title = "On improving random forest for hard-to-classify records",

abstract = "Random Forest draws much interest from the research community because of its simplicity and excellent performance. The splitting attribute at each node of a decision tree for Random Forest is determined from a predefined number of randomly selected subset of attributes of the entire attribute set. The size of the subset is one of the most controversial points of Random Forest that encouraged many contributions. However, a little attention is given to improve Random Forest specifically for those records that are hard to classify. In this paper, we propose a novel technique of detecting hard-to-classify records and increase the weights of those records in a training data set. We then build Random Forest from the weighted training data set. The experimental results presented in this paper indicate that the ensemble accuracy of Random Forest can be improved when applied on weighted training data sets with more emphasis on hard-to-classify records.",

keywords = "Decision forest, Randam Forest, Ensemble accuracy",

author = "Adnan, {Md Nasim} and Islam, {Md Zahidul}",

note = "Imported on 03 May 2017 - DigiTool details were: publisher = 2016. Event dates (773o) = December 12 - 15, 2016; Parent title (773t) = International Conference on Advanced Data Mining and Applications (ADMA).; Advanced Data Mining and Applications (ADMA) 12th International Conference ; Conference date: 12-12-2016 Through 15-12-2016",

year = "2016",

doi = "10.1007/978-3-319-49586-6_39",

language = "English",

isbn = "9783319495859",

volume = "10086",

pages = "558--566",

editor = "Jinyan Li and Xue Li and Shuliang Wang and Jianxin Li and Sheng, {Quan Z.}",

booktitle = "Proceedings of the 12th International Conference on Advanced Data Mining and Applications, ADMA 2016",

publisher = "Springer",

address = "United States",

url = "https://cs.adelaide.edu.au/~adma2016/, https://cs.adelaide.edu.au/~adma2016/ADMA2016_program.pdf",

}

Adnan, MN & Islam, MZ 2016, On improving random forest for hard-to-classify records. in J Li, X Li, S Wang, J Li & QZ Sheng (eds), Proceedings of the 12th International Conference on Advanced Data Mining and Applications, ADMA 2016. vol. 10086, Springer, Switzerland, pp. 558-566, Advanced Data Mining and Applications (ADMA) 12th International Conference, Gold Coast, Queensland, Australia, 12/12/16. https://doi.org/10.1007/978-3-319-49586-6_39

On improving random forest for hard-to-classify records. / Adnan, Md Nasim; Islam, Md Zahidul.
Proceedings of the 12th International Conference on Advanced Data Mining and Applications, ADMA 2016. ed. / Jinyan Li; Xue Li; Shuliang Wang; Jianxin Li; Quan Z. Sheng. Vol. 10086 Switzerland: Springer, 2016. p. 558-566.

Research output: Book chapter/Published conference paper › Conference paper › peer-review

TY - GEN

T1 - On improving random forest for hard-to-classify records

AU - Adnan, Md Nasim

AU - Islam, Md Zahidul

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = 2016. Event dates (773o) = December 12 - 15, 2016; Parent title (773t) = International Conference on Advanced Data Mining and Applications (ADMA).

PY - 2016

Y1 - 2016

N2 - Random Forest draws much interest from the research community because of its simplicity and excellent performance. The splitting attribute at each node of a decision tree for Random Forest is determined from a predefined number of randomly selected subset of attributes of the entire attribute set. The size of the subset is one of the most controversial points of Random Forest that encouraged many contributions. However, a little attention is given to improve Random Forest specifically for those records that are hard to classify. In this paper, we propose a novel technique of detecting hard-to-classify records and increase the weights of those records in a training data set. We then build Random Forest from the weighted training data set. The experimental results presented in this paper indicate that the ensemble accuracy of Random Forest can be improved when applied on weighted training data sets with more emphasis on hard-to-classify records.

AB - Random Forest draws much interest from the research community because of its simplicity and excellent performance. The splitting attribute at each node of a decision tree for Random Forest is determined from a predefined number of randomly selected subset of attributes of the entire attribute set. The size of the subset is one of the most controversial points of Random Forest that encouraged many contributions. However, a little attention is given to improve Random Forest specifically for those records that are hard to classify. In this paper, we propose a novel technique of detecting hard-to-classify records and increase the weights of those records in a training data set. We then build Random Forest from the weighted training data set. The experimental results presented in this paper indicate that the ensemble accuracy of Random Forest can be improved when applied on weighted training data sets with more emphasis on hard-to-classify records.

KW - Decision forest

KW - Randam Forest

KW - Ensemble accuracy

U2 - 10.1007/978-3-319-49586-6_39

DO - 10.1007/978-3-319-49586-6_39

M3 - Conference paper

SN - 9783319495859

VL - 10086

SP - 558

EP - 566

BT - Proceedings of the 12th International Conference on Advanced Data Mining and Applications, ADMA 2016

A2 - Li, Jinyan

A2 - Li, Xue

A2 - Wang, Shuliang

A2 - Li, Jianxin

A2 - Sheng, Quan Z.

PB - Springer

CY - Switzerland

T2 - Advanced Data Mining and Applications (ADMA) 12th International Conference

Y2 - 12 December 2016 through 15 December 2016

ER -

On improving random forest for hard-to-classify records

Abstract

Conference

Access to Document

Fingerprint

Cite this