Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
 
Articles by T. Santhanam
Total Records ( 14 ) for T. Santhanam
  B.V. Sumana and T. Santhanam
  Dataset with class imbalance is a challenging problem in many real-world application domains in the field of machine learning and data mining community which is the main cause for the degradation of the classifier performance. A data set is said to be imbalanced if the distribution of instances belonging to each class is not in equal proportion. Researchers worked on class imbalance have identified that combination of class overlapping with class imbalance and high dimensional data is crucial problems which are the important factors for the deterioration of the classifier performance. To overcome this problem a model with two phases of preprocessing is proposed. The objective of the proposed model is 3 fold: increase the minority class instances to address the class imbalance problem, to remove class overlap using the proposed model and to reduce Type 1 and 2 error of the classifier, i.e., false positive and false negative rate which means that the patients who actually does not have disease but predicted as to have disease and vice versa which is a serious problem in reality as it is a matter of life of a patient. The efficiency of the proposed model was evaluated with performance measures like precision, recall, F-measure, AUC, accuracy, kappa, false positive rate and false negative rate. Results proved that proposed model is more efficient than the existing models in the literature as all the 9 classifiers on all the three datasets showed accuracy above 99% and a significant reduction in false positive and false negative rate and also the proposed model was successful in overcoming the issues associated with the real world data sets like class overlap and class imbalance and finally improving the performances of the classifier like false positive, false negative rate, Auc and accuracy of the classifier.
  A. Punitha and T. Santhanam
  One of the major challenges in medical domain is the extraction of intelligible knowledge from medical diagnosis data. It is quite common among the researching community to apply Principal Component Analysis (PCA) for the extraction of prominent features and to use feature correlation method for redundant features removal. This paper discusses a three-phase approach selection technique to extract features for further usage in clinical practice for better understanding and prevention of superfluous medical events. In the first phase PCA is employed to extract the relevant features followed by the elimination of redundant features using the class correlation and feature correlation technique in phase two and in the final phase Learning Vector Quantization (LVQ) network is utilized for classification. The proposed method is validated upon Wisconsin Breast Cancer Database (WBCD), which is a very well known dataset obtained from the UCI machine-learning repository. The abridged feature set and classification accuracy are found to be satisfactory.
  Mary Metilda and T. Santhanam
  The task of detecting faces is a precursor intended for shaping the information that the face provides. A robust way to locate the faces in images, insensitive to scale, pose, style, facial expression and lighting condition is contributed as a two phase process in this study. The first phase aspects a human face against a complex background by performing a skin color analysis of the image, as it is most liable that color noises have low probability on being flesh color. During the second phase, random measurements are generated to populate the entrant face region according to the face anthropometrics using the eye location determined from the YCbCr color space. The combination of holistic and configural approach implemented in this study has proven to be a very speedy, effortless and exceedingly competent scheme for locating complex stimuli such as face.
  C. Jayakumari and T. Santhanam
  A novel technique of intelligent segmentation and classification of exudates for diabetic retinopathy by applying energy minimization method using a recurrent neural network that is an Echo State Neural Network (ESNN) which, yields highly satisfactory results when compared with that of an existing contextual clustering segmentation (CC) is explored in this study. The modular neural network is trained using a set of 30 images consisting of 5 normal images and 25 abnormal images. The trained system has been tested with 5 normal and 20 abnormal images and is found to acquire satisfactory results with 90% (18/20) sensitivity.
  M. Nachamai , T. Santhanam and M. Muthuraman
  This study proposes a methodology to find the interest levels of two speakers in a conversation. The ANN-HMM approach-a hybrid method is adopted. The hybrid method uses language input as an additional parameter in addition to the acoustic features. The language input provides a measure of classification of the input speech utterance. A combined classifier is used to make a linear decision on the emotion of the uttered speech as an arousal or valence. When the decision is fed to the Generative Factor Analyzed Hidden Markov Model (GFA-HMM) it evidently substantiates to be a better method with good accuracy rate of classification of whether the speaker is entangled in the conversation or vice-versa. The proposed method produced highly satisfactory results for the Linguistic Data Consortium (LDC) emotional prosody dataset.
  T. Velmurugan and T. Santhanam
  Clustering is one of the most important research areas in the field of data mining. Clustering means creating groups of objects based on their features in such a way that the objects belonging to the same groups are similar and those belonging in different groups are dissimilar. Clustering is an unsupervised learning technique. Data clustering is the subject of active research in several fields such as statistics, pattern recognition and machine learning. From a practical perspective clustering plays an outstanding role in data mining applications in many domains. The main advantage of clustering is that interesting patterns and structures can be found directly from very large data sets with little or none of the background knowledge. Clustering algorithms can be applied in many areas, for instance marketing, biology, libraries, insurance, city-planning, earthquake studies and www document classification. Data mining adds to clustering the complications of very large datasets with very many attributes of different types. This imposes unique computational requirements on relevant clustering algorithms. A variety of algorithms have recently emerged that meet these requirements and were successfully applied to real-life data mining problems. They are subject of this survey. Also, this survey explores the behavior of some of the partition based clustering algorithms and their basic approaches with experimental results.
  B. Sujatha and T. Santhanam
  Lipreading a perception of speech for listeners with hearing impairment is purely based on observing the lip movements under noisy conditions where visual speech information plays an important role. Lipreading, a visual modality which involves watching the movement of lips constitutes 1/3 of the conveyed message. This study investigates the use of two feature extraction methodologies for recognizing isolated words. The first type is based on a geometric approach which extracts the features like inner height, inner width, outer height and outer width of the lips while the second type is based on a set of Block-Based Gabor-Wavelet co-efficient extracted from each frame. Then these two features are given as input into the Ergodic Hidden Markov Model for recognizing the words.
  K. Meenakshi Sundaram , T. Santhanam , M. Saroja and C. P. Sumathi
  Problem statement: An important tool in the field of education methodology is examination. As far as teaching-learning-evaluation process is concerned, the major task associated with the objective-type examination system is the administration of question paper setting. The lack of expertise and time are the major constraints that are encountered in the task of setting objective type question papers. During retrieval of records from a objective type question bank, redundancy may occur. To solve this problem, an approach needed to retrieve records from a database without redundancy. Approach: The task associated in generating the required collection of questions from a question bank, with minimal redundancy as far as possible in the retrieval of records from the question bank using mid-square and mid-product techniques for random number generation were discussed in this study. Results: A modified approach was identified and handled to generate random numbers and used to retrieve records from a database Conclusion: The suggested modified approach was more suitable for retrieving records from a database of even smaller size.
  T. Santhanam and S. Radhika
  Problem statement: Noise identification is predominant in any digital image processing algorithms, which helps in identifying the filters to smooth the image for further processing. Approach: An Artificial Neural Network (ANN) based approach was proposed for noise identification. The suggested technique involved seclusion of the noise samples and extracts their statistical features, which was then applied to a neural network to identify the noise. Results: Neural networks provided a better solution in identifying the noise. Conclusion: By identifying the noise the precise filter can be used for enhancing the given image.
  T. Santhanam and Shyam Sundaram
  Problem statement: This study used data mining modeling techniques to examine the blood donor classification. The availability of blood in blood banks is a critical and important aspect in a healthcare system. Blood banks (in the developing countries context) are typically based on a healthy person voluntarily donating blood and is used for transfusions or made into medications. The ability to identify regular blood donors will enable blood banks and voluntary organizations to plan systematically for organizing blood donation camps in an effective manner. Approach: Identify the blood donation behavior using the classification algorithms of data mining. The analysis had been carried out using a standard blood transfusion dataset and using the CART decision tree algorithm implemented in Weka. Results: Numerical experimental results on the UCI ML blood transfusion data with the enhancements helped to identify donor classification. Conclusion: The CART derived model along with the extended definition for identifying regular voluntary donors provided a good classification accuracy based model.
  M. Gopikrishnan and T. Santhanam
  Problem statement: A biometric system provides automatic identification of an individual based on a unique feature or characteristic possessed by the individual. Iris recognition is regarded as the most reliable and accurate biometric identification system available. Approach: Most commercial iris recognition systems use patented algorithms developed by Daugman and these algorithms are able to produce perfect recognition rates. However, published results have usually been produced under favorable conditions and there have been no independent trials of the technology. Results: In this study after providing brief picture on development of various techniques for iris recognition, hamming distance coupled with neural network based iris recognition techniques were discussed. Perfect recognition on a set of 150 eye images has been achieved through this approach. Further, Tests on another set of 801 images resulted in false accept and false reject rates of 0.0005 and 0.187% respectively, providing the reliability and accuracy of the biometric technology. Conclusion/Recommendations: This study provided results of iris recognition performed applying Hamming distance, Feed forward back propagation, Cascade forward back propagation, Elman forward back propagation and perceptron. It has been established that the method suggested applying perceptron provides the best accuracy in respect of iris recognition with no major additional computational complexity.
  T. Velmurugan and T. Santhanam
  Problem statement: Clustering is one of the most important research areas in the field of data mining. Clustering means creating groups of objects based on their features in such a way that the objects belonging to the same groups are similar and those belonging to different groups are dissimilar. Clustering is an unsupervised learning technique. The main advantage of clustering is that interesting patterns and structures can be found directly from very large data sets with little or none of the background knowledge. Clustering algorithms can be applied in many domains. Approach: In this research, the most representative algorithms K-Means and K-Medoids were examined and analyzed based on their basic approach. The best algorithm in each category was found out based on their performance. The input data points are generated by two ways, one by using normal distribution and another by applying uniform distribution. Results: The randomly distributed data points were taken as input to these algorithms and clusters are found out for each algorithm. The algorithms were implemented using JAVA language and the performance was analyzed based on their clustering quality. The execution time for the algorithms in each category was compared for different runs. The accuracy of the algorithm was investigated during different execution of the program on the input data points. Conclusion: The average time taken by K-Means algorithm is greater than the time taken by K-Medoids algorithm for both the case of normal and uniform distributions. The results proved to be satisfactory.
  B. Sujatha and T. Santhanam
  Problem statement: Deaf and dumb needs assistance from a technical box that takes movements of lips to identify the words. This technical article provided appropriate model implementation of flexible lip model for better visual lip reading system. Approach: From the frame sequence of words, Active Shape Model (ASM) based lip model provided local tracking and extraction of geometric lip-feature. Two geometric criteria define required geometric features and its variations in the sequence. Results: The feature established machine classification using Analytic Hierarchy Process (AHP), a relative weight finder. AHP presents weight vector to fuzzy classifier to decide the video frame sequence belonging to a respective word. Conclusion: The suggested model tested on a total of 5 different sample databases results in 83.2% accuracy over the other combinational algorithms.
  C. P. Sumathi , R. Padmaja Valli and T. Santhanam
  Problem statement: With the continued growth and proliferation of e-commerce, Web services and Web-based information systems, the volumes of click-stream and user data collected by Web-based organizations in their daily operations have reached astronomical proportions. Analyzing such data can help these organizations optimize the functionality of web-based applications and provide more personalized content to visitors. This type of analysis involved the automatic discovery of usage interest on the web pages which are often stored in web and applications server access logs. Approach: The usage interest on the web pages in various sessions was partitioned into clusters such that sessions with "similar" interest were placed in the same cluster using expectation maximization clustering technique as discussed in this study. Results: The approach results in the generation of usage profiles and automatic identification of user interest in each profile. Conclusion: The significance of the results will be helpful for organizations for web site improvement based on their navigational interest and provide recommendations for page(s) not yet visited by the user.
 
 
 
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility