|
|
Articles
by
S.P. Rajagopalan |
Total Records (
13 ) for
S.P. Rajagopalan |
|
 |
|
|
|
T. Muralidharan
,
V. Saishanmuga Raja
and
S.P. Rajagopalan
|
|
In the most recent couple of years as internet utilization turns into the principle supply route of the
lifes every day exercises, the issue of spam turns out to be intense for web group. Spam pages frame a genuine
risk for a wide range of clients. This risk demonstrated to advance constantly with no piece of information to
lessen. Diverse types of spam saw an emotional increment in both size and negative effect. A lot of e-mails and
website pages are considered spam either in Simple Mail Transfer Protocol (SMTP) or web crawlers. Numerous
specialized strategies were proposed to approach the issue of spam. We propose a Hybrid Extensive Machine
Learning Algorithm (HEMLA) for detection and classification of that offers weight to the data nourished by
clients and thinks about the presence of some space particular components. Hybrid extensive machine learning
algorithm is a combination of many learning algorithms like conjugate gradient, resilient back-propagation and
levenberg-marquardt algorithms. The outcomes demonstrate that the hybrid extensive machine learning
algorithm overcomes the traditional web filtering methods as far as reducing the false positives and the false
negatives and increasing the accuracy. |
|
|
|
|
S. Mohandoss
,
V. Sai Shanmuga Raja
and
S.P. Rajagopalan
|
|
Cardiovascular illness remains the greatest reason for deaths worldwide and the heart disease
prediction at the early stage is significance. In this study, we propose a hybrid heart disease prediction system
using evolutionary learning algorithms like cascaded neural network and Genetic algorithm. It is used for heart
disease prediction at the early stage utilizing the patients therapeutic record. The results are compared with
the known supervised classifier Support Vector Machine (SVM). During classification, 13 attributes are given
as input to the CNN classifier to predict the risk of heart illness. The proposed framework can be used as a guide
by the doctors to predict the disease in a more productive way. The effectiveness of the classifier is tried
utilizing the records gathered from 270 patients. The outcomes demonstrate that the Genetic based CNN
classifier can anticipate the probability of patients with coronary illness in a more effective manner. |
|
|
|
|
D. Elantamilan
,
V. SaiShanmuga Raja
and
S.P. Rajagopalan
|
|
In this research, we have presented a technique for individual recognizable proof in view of iris
recognition utilizing genetic algorithm and neural network. The procedure of iris recognition comprises of
confinement of the iris locale and area of information set of iris pictures took after by iris design recognition.
A neural network is utilized to diminish the low recognition rate, low accuracy and expanded time of
recuperation. Here, the genetic algorithm is utilized to upgrade the neural networks parameters. The reenactment
comes about demonstrate a decent recognizable proof rate and lessened preparing time. The iris became
a much-explored field. Human iris contains unique and very important information about persons. |
|
|
|
|
|
R. Radha
and
S.P. Rajagopalan
|
|
Fuzzy logic is a computational paradigm that provides a mathematical tool for dealing with the uncertainty and the imprecision typical of human reasoning. A prime characteristic of fuzzy logic is its capability of expressing knowledge in a linguistic way, allowing a system to be described by simple, human-friendly rules. The fuzzy set framework has been utilized in several different approaches to modeling the diagnostic process. In this paper Diabetes related diseases and their symptoms are taken. The physician’s medical knowledge is represented as a fuzzy relation between symptoms and diseases. Thus, given the fuzzy set A of the symptoms observed in the patient and the fuzzy relation R representing the medical knowledge that relates the symptoms in set S to the diseases in set D, then the fuzzy set B of the possible diseases of the patient can be inferred by means of the compositional rule of inference. Fuzzy membership values for representing different symptoms are framed and they are used for forming the relations. |
|
|
|
|
S. Jayalakshmi
and
S.P. Rajagopalan
|
|
In this research, a parallel implementation of a Modular Simulated Annealing (MSA) algorithm, applied to classical Job-Shop Scheduling (JSS) problems is presented. The implementation has been done as a multiple island system suitable to run on the Distributed Resource Machine (DRM) environment, which is a novel scalable, distributed virtual machine developed based on Java technology. The support of the DRM environment was very effective with respect to message passing, having collaboration with a remote machine. The empirical results show that the method proposed is quite successful compared to the ordinary MSA and other systems described in literature. |
|
|
|
|
B. Venkata Raju
and
S.P. Rajagopalan
|
|
The study on expert systems led to the identification of the ‘knowledge acquisition bottleneck, that it was generally extremely difficult to make overt the presumed knowledge of human experts in order to program it for computers. The history and reasons for the adoption of repertory grid methodologies and tools to overcome the knowledge acquisition bottleneck are described. Then a more fundamental analysis is made of why expert systems to date have had only limited success and merits of a personal construct approach to emulating human expertise in greater depth than has been achieved with existing cognitive science models are presented. In conclusion, it is noted that the techniques developed to emulate human expertise are essentially ones for modeling and emulating any persons psychological processes, not just those of people valued by others as experts. PCP-based expert systems methods and technology have wide relevance, for example, in clinical and educational research and applications. The role of personal construct psychology in computer research and applications concerned with the development of expert systems and their beginnings in artificial intelligence and cognitive science are covered in this study. |
|
|
|
|
D. Pugazhenthi
and
S.P. Rajagopalan
|
|
Drug discovery refers to the finding of a new drug which could be a completely new compound or a new derivative of existing compounds. Drug discovery is the ultimate goal of drug design which concerned with the design of a chemical compound that exhibits a desired pharmacological activity. Machine learning tools, in particular Support Vector Machines (SVM), Particle Swarm Optimisation (PSO) and Genetic Programming (GP), are increasingly used in pharmaceuticals research and development. They are inherently suitable for use with noisy, high dimensional data, as is commonly used in cheminformatic, bioinformatics and other types of drug research studies. These aspects are demonstrated via review of their current usage and future prospects in context with drug discovery activities. |
|
|
|
|
M. Sundara Rajan
and
S.P. Rajagopalan
|
|
It has often been thought that word sense ambiguity
is a cause of poor performance in Information Retrieval (IR) systems.
The belief is that if ambiguous words can be correctly disambiguated,
IR performance will increase. However, recent research into the application
of a word sense disambiguator to an IR system failed to show any performance
increase. From these results it has become clear that more basic research
is needed to investigate the relationship between sense ambiguity, disambiguation
and IR. Using a technique that introduces additional sense ambiguity into
a collection, this study presents research that goes beyond previous work
in this field to reveal the influence that ambiguity and disambiguation
have on a probabilistic IR system. We conclude that word sense ambiguity
is only problematic to an IR system when it is retrieving from very short
queries. In addition we argue that if a word sense disambiguator is to
be of any use to an IR system, the disambiguator must be able to resolve
word senses to a high degree of accuracy. |
|
|
|
|
R. Radha
,
S. Jayalakshmi
and
S.P. Rajagopalan
|
|
In this study we show that the gene-expression
profiles based on microarray analysis can be used to predict the genes
which are responsible for death of the patient in early-stage B-cell Lymphoma.
This study presents a new computational method for this prediction and
use an informative subset of samples to identify genes whose expression
is related to the problem under study. A reduction technique integrating
global normalization, fuzzy membership generation, Pearson`s correlation
method are used to identify genes which are related to non survival. The
early identification of the genes causing B-cell Lymphoma that leads to
death will benefit patients in the sense that the patients can be provided
proper therapy to increase their life span. The validity of the results
are established with the help of statistical methods. |
|
|
|
|
K. Shyamala
and
S.P. Rajagopalan
|
|
The main objective of any higher educational institution is to impart quality education. One way to reach the highest level of quality in higher education systems is by improving the decision making procedures on various processes such as assessment, evaluation, counseling and so on which requires knowledge. The knowledge is hidden among the educational data set and it is extractable through data mining technology. This paper is designed to present and justify the capabilities of data mining in the context of higher education by offering a data mining model for higher educational system in the colleges. It presents an approach to classifying students in order to predict their final grade based on certain features extracted from educational data bases. It helps earlier in identifying the dropouts and students who are below average and allow the teacher to provide appropriate counseling/advising in appropriate time. |
|
|
|
|
|
V.N. Rajavarman
and
S.P. Rajagopalan
|
|
We discovered genetic features and environmental factors which were involved in multifactorial diseases. To exploit the massive data obtained from the experiments conducted at the General Hospital, Chennai, data mining tools were required and we proposed a 2-Phase approach using a specific genetic algorithm. This heuristic approach had been chosen as the number of features to consider was large (upto 3654 for biological data under our study). Collected data indicated for pairs of affected individuals of a same family their similarity at given points (locus) of their chromosomes. This was represented in a matrix where each locus was represented by a column and each pairs of individuals considered by a row. The objective was first to isolate the most relevant associations of features and then to class individuals that had the considered disease according to these associations. For the first phase, the feature selection problem, we used a genetic algorithm (GA). To deal with this very specific problem, some advanced mechanisms had been introduced in the genetic algorithm such as sharing, random immigrant, dedicated genetic operators and a particular distance operator had been defined. Then, the second phase, a clustering based on the features selected during the previous phase, will use the clustering algorithm k-means. |
|
|
|
|
D. Pugazhenthi
and
S.P. Rajagopalan
|
|
Problem statement: Activities of drug molecules can be predicted by Quantitative Structure Activity Relationship (QSAR) models, which overcome the disadvantage of high cost and long cycle by employing traditional experimental methods. With the fact that number of drug molecules with positive activity is rather fewer than that with negatives, it is important to predict molecular activities considering such an unbalanced situation. Approach: Asymmetric bagging and feature selection was introduced into the problem and Asymmetric Bagging of Support Vector Machines (AB-SVM) was proposed on predicting drug activities to treat unbalanced problem. At the same time, features extracted from structures of drug molecules affected prediction accuracy of QSAR models. Hybrid algorithm named SPRAG was proposed, which applied an embedded feature selection method to remove redundant and irrelevant features for AB-SVM. Results: Numerical experimental results on a data set of molecular activities showed that AB-SVM improved AUC and sensitivity values of molecular activities and SPRAG with feature selection further helps to improve prediction ability. Conclusion: Asymmetric bagging can help to improve prediction accuracy of activities of drug molecules, which could be furthermore improved by performing feature selection to select relevant features from the drug. |
|
|
|
|
|
J. Sujatha
and
S.P. Rajagopalan
|
|
This study proposes a method to identify Parkinson disease classifying patients as normal or
abnormal using the latest machine learning algorithms. The image is acquired, converted to gray scale,
preprocessed using Wiener filter. Canny edge detection method is used which involves image smoothing,
gradient operation, non maxima suppression, hysteresis thresholding and connectivity analysis. Then image
is segmented using fuzzy C-means. Features are extracted using GLCM technique and ANFIS classification is
used to classify patients as normal or abnormal. Experimental results proves that patients suffering from
neurological disease can be effectively detected using this method. A total of 167 spiral images were used out
of which 56 were normal patient and 111 were abnormal collected from various sources. A classification
accuracy of 99% is achieved. |
|
|
|
|
|
|