Kongu Engineering College

About: Kongu Engineering College is a based out in . It is known for research contribution in the topics: Computer science & Cluster analysis. The organization has 2001 authors who have published 1978 publications receiving 16923 citations.

A Hash based Mining Algorithm for Maximal Frequent Item Sets using Linear Probing

Abstract: Data mining is having a vital role in many of the applications like market-basket analysis, in biotechnology field etc. In data mining, frequent itemsets plays an important role which is used to identify the correlations among the fields of database. In this paper, we propose an algorithm, HBMFI-LP which hashing technology to store the database in vertical data format. To avoid hash collisions, linear probing technique is utilized. The proposed algorithm generates the exact set of maximal frequent itemsets directly by removing all nonmaximal itemsets. The proposed algorithm is compared with the recently developed MAFIA algorithm and is shown that the HBMFI-LP outperforms in the order of two to three

Informative Gene Selection for Cancer Classification with Microarray Data Using a Metaheuristic Framework

Abstract: Objective: Cancer diagnosis is one of the most vital emerging clinical applications of microarray data. Due to the high dimensionality, gene selection is an important step for improving expression data classification performance. There is therefore a need for effective methods to select informative genes for prediction and diagnosis of cancer. The main objective of this research was to derive a heuristic approach to select highly informative genes. Methods: A metaheuristic approach with a Genetic Algorithm with Levy Flight (GA-LV) was applied for classification of cancer genes in microarrays. The experimental results were analyzed with five major cancer gene expression benchmark datasets. Result: GA-LV proved superior to GA and statistical approaches, with 100% accuracy for the dataset for Leukemia, Lung and Lymphoma. For Prostate and Colon datasets the GA-LV was 99.5% and 99.2% accurate, respectively. Conclusion: The experimental results show that the proposed approach is suitable for effective gene selection with all benchmark datasets, removing irrelevant and redundant genes to improve classification accuracy.

Computational wear assessment of hard on hard hip implants subject to physically demanding tasks

Abstract: Hip implants subject to gait loading due to occupational activities are potentially prone to failures such as osteolysis and aseptic loosening, causing painful revision surgeries. Highly risky gait activities such as carrying a load, stairs up or down and ladder up or down may cause excessive loading at the hip joint, resulting in generation of wear and related debris. Estimation of wear under the above gait activities is thus crucial to design and develop a new and improved implant component. With this motivation, this paper presents an assessment of wear generation of PCD-on-PCD (poly crystalline diamond) hip implants using finite element (FE) analysis. Three-dimensional (3D) FE model of hip implant along with peak gait and peak flexion angle for each activity was used to estimate wear of PCD for 10 million cycles. The maximum and minimum initial contact pressures of 206.19 MPa and 151.89 MPa were obtained for carrying load of 40 kg and sitting down or getting up activity. The simulation results obtained from finite element model also revealed that the maximum linear wear of 0.585 μm occurred for the patients frequently involved in sitting down or getting up gait activity and maximum volumetric wear of 0.025 mm3 for ladder up gait activity. The stair down activity showed the least linear and volumetric wear of 0.158 μm and 0.008 mm3, respectively, at the end of 10 million cycles. Graphical abstract Computational wear assessment of hip implants subjected to physically demanding tasks.

A novel approach for imputation of missing continuous attribute values in databases using genetic algorithm

Abstract: Missing values in databases are more common and if untreated distort the estimates. Numerous methods were developed by researchers to replace the missing values in continuous attributes. The simple methods used are less efficient and the efficient methods are very complex to implement. Hence, to maintain a balance between simplicity and efficiency a new method called Bayesian genetic algorithm BGA is proposed based on genetic algorithm and Bayes theorem for both missing at random MAR and missing completely at random MCAR assumption. Accuracy of BGA is compared with that of mean, kNN and multiple imputation in finding the missing values and the results are studied. BGA produces more accurate results than other methods in four datasets studied at different rates of missingness ranging from 5% to 60%. BGA works better even in large datasets resulting in less biased estimates.

An enhanced data mining model for text classification

Abstract: Classification plays a vital role in many information management and retrieval tasks. This paper studies classification of text document. Text classification is a supervised technique that uses labeled training data to learn the classification system and then automatically classifies the remaining text using the learned system. In this paper, we propose a mining model consists of sentence-based concept analysis, document-based concept analysis, and corpus-based concept-analysis. Then we analyze the term that contributes to the sentence semantics on the sentence, document, and corpus levels rather than the traditional analysis of the document only. After extracting feature vector for each new document, feature selection is performed. It is then followed by K-Nearest Neighbour classification. The approach enhances the text classification accuracy.