Privacy preserving data mining using particle swarm optimisation trained auto-associative neural network: an application to bankruptcy prediction in banks
by Paramjeet; V. Ravi; Naveen Nekuri; Chillarige Raghavendra Rao
International Journal of Data Mining, Modelling and Management (IJDMMM), Vol. 4, No. 1, 2012

Abstract: While data mining made inroads into the diverse areas it also entails violation of individual privacy leading to legal complications in areas like medicine and finance as consequently, privacy preserving data mining (PPDM) emerged as a new area. To achieve an equitable solution to this problem, data owners must not only preserve privacy and but also guarantee valid data mining results. This paper proposes a novel particle swarm optimisation (PSO) trained auto associative neural network (PSOAANN) for privacy preservation. Then, decision tree and logistic regression are invoked for data mining purpose, leading to PSOAANN + DT and PSOAANN + LR hybrids. The efficacy of hybrids is tested on five benchmark and four bankruptcy datasets. The results are compared with those of Ramu and Ravi (2009) and others. It was observed that the proposed hybrids yielded better or comparable results. We conclude that PSOAANN can be used as viable approach for privacy preservation.

Online publication date: Sat, 23-Aug-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Data Mining, Modelling and Management (IJDMMM):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com