Materials & Design Volume 28, Issue 8, 2007, Pages 2379-2386

doi:10.1016/j.matdes.2006.07.018 | How to Cite or Link Using DOI Copyright 2006 Elsevier Ltd All rights reserved.

A Nave-Bayes classifier for damage detection in engineering materials

O. Addina, S.M. Sapuanb, , , E. Mahdic and M. Othmand
a

Laboratory of Intelligent Systems, Institute of Advanced Technology, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
b

Department of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
c

Department of Aerospace Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia

Department of Communication Technology and Networks, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia Received 26 January 2006; accepted 19 July 2006. Available online 29 September 2006.

Abstract
This paper is intended to introduce the Bayesian network in general and the Nave-Bayes classifier in particular as one of the most successful classification systems to simulate damage detection in engineering materials. A method for feature subset selection has also been introduced too. The method is based on mean and maximum values of the amplitudes of waves after dividing them into folds then grouping them by a clustering algorithm (e.g. k-means algorithm). The Nave-Bayes classifier and the feature sub-set selection method were analyzed and tested on two sets of data. The data sets were conducted based on artificial damages created in quasi isotopic laminated composites of the AS4/3501-6 graphite/epoxy system and ball bearing of the type 6204 with a steel cage. The NaveBayes classifier and the proposed feature subset selection algorithm have been shown as efficient techniques for damage detection in engineering materials.

http://www.experiencefestival.com/wp/videos/lecture-13--machine-learningstanford/LBtuYU-HfUg

Materials & Design Volume 32, Issue 1, January 2011, Pages 462-467

doi:10.1016/j.matdes.2010.05.055 | How to Cite or Link Using DOI Copyright 2010 Elsevier Ltd All rights reserved.

Permissions & Reprints

Technical Report

Prediction of tensile strength of friction stir welded aluminium matrix TiCp particulate reinforced composite
S. Gopalakrishnana,
a , , ,

and N. Muruganb, 1,

Department of Mechanical Engineering, K.S. Rangasamy College of Technology, Tiruchengode 637 215, Tamil Nadu, India
b

Department of Mechanical Engineering, Coimbatore Institute of Technology, Coimbatore 641 014, Tamil Nadu, India Received 11 February 2010; accepted 28 May 2010. Available online 2 June 2010.

Abstract
The usage of particulate reinforced metal matrix composite (MMC) is steadily increasing due to its properties such as high specific strength, high specific modulus and good wear resistance. Aluminium matrix composite (AMC) plays an important role to meet the above requirements. Effective utilization of AMC is based on not only its production but also on fabrication methods. Among AMCs, those based on particulate reinforcements are particularly attractive, due to their lower production costs. Aluminium matrix titanium carbide reinforced composite (AlTiCp) was produced in an inert atmosphere by indigenously developed Modified Stir Casting Process with bottom pouring arrangement (37% TiC by weight). Friction stir welding process (FSW) is employed to make weld joints. The welding parameters such as axial force, welding speed, tool rotational speed, percentage TiC addition etc., and profile of the tool were considered for analysis. In this study, an attempt is made to predict ultimate tensile strength (UTS) of the welded joints using a mathematical model. The FSW specimens without any post-weld heat treatment belonging to a different set of parameters tested, exhibited a high joint efficiency (most of them ranging from 90% to 98%) with respect to the ultimate tensile strength of the base material AA6061. It was found from the analysis of the model that the tool pin profile and the welding speed have more significant effect on tensile strength.

ntern ational Journal of Soft Computing

Year: 2007 | Volume: 2 | Issue: 4 | Page No.: 488-493 DOI: 10.3923/ijscomp.2007.488.493

A Novel Approach to Concept Extraction Using Naive Bayesian Classification Technique

M. Sathya , P.Venil and M.S. Saleem Basha

Abstract: Most of the information resources are capable to provide concept dependent results for the biased query. But the retrieval mechanism could not provide the relevant documents for the query text due to the size of the information resources is dynamically growing as the new topics being added. This problem can be overcome by automatically generating wrappers for these hidden documents. We are proposing a novel approach for automatically generating wrappers for describing the content of the hidden documents using a co-occurrence based clustering algorithm and Naive Bayesian classification model. The initial stage is the learning stage, which clusters the document based on the distinct concepts present in that. The learning technique makes use of a thesaurus and builds a co-occurrence correlation model. Then the clustered document features are used to generate the concept description using Naive Bayesian classifier. The join and posterior probabilities are calculated using the greedy selection and joining algorithm to represent cluster. Our implementation was tested on the standard data set and shows a better performance. How to cite this article: Sathya, M., M.S.S. Basha and P. Venil, 2007. A novel approach to concept extraction using naive bayesian classification technique. Int. J. Soft Comput., 2: 488-493. DOI: 10.3923/ijscomp.2007.488.493 URL: http://medwelljournals.com/abstract/?doi=ijscomp.2007.488.493

Information Systems Volume 21, Issue 3, May 1996, Pages 269-298

doi:10.1016/0306-4379(96)00015-4 | How to Cite or Link Using DOI Copyright 1996 Published by Elsevier Science Ltd.

Permissions & Reprints

Selection predicate indexing for active databases using interval skip lists*1
Eric N. Hanson and Theodore Johnson Computer and Information Science and Engineering Department, University of Florida, Gainesville, FL 32611-6120, USA Received 1 June 1995; revised 20 April 1996. Available online 9 February 1999.

Abstract

A new, efficient selection predicate indexing scheme for active database systems is introduced. The selection predicate index proposed uses an interval index on an attribute of a relation or object collection when one or more rule condition clauses are defined on that attribute. The selection predicate index uses a new type of interval index called the interval skip list (IS-list). The IS-list is designed to allow efficient retrieval of all intervals that overlap a point, while allowing dynamic insertion and deletion of intervals. IS-list algorithms are described in detail. The IS-list allows efficient on-line searches, insertions, and deletions, yet is much simpler to implement than other comparable interval index data structures such as the priority search tree and balanced interval binary search tree (IBS-tree). IS-lists require only one third as much code to implement as balanced IBS-trees. The combination of simplicity, performance, and dynamic updateability of the IS-list is unmatched by any other interval index data structure. This makes the IS-list a good interval index structure for implementation in an active database predicate index. Author Keywords: Active Databases; Triggers; Predicate Indexing; Interval Skip Lists; Discrimination Networks