Manas Ranjan Senapati

Mining for similarities in time series data using wavelet-based feature vectors and neural networks

This paper presents a comparison between different wavelet feature vectors for data mining of nonstationary time series that occurs in an electricity supply network. Three different wavelet algorithms are simulated and applied on nine classes of power signal time series, which primarily belongs to an important problem area called electric power quality. In contrast to the wavelet analysis, the paper presents a new approach called S-transform-based time frequency analysis in processing power quality disturbance data. Certain pertinent feature vectors are extracted using the well-known wavelet methods and the new approach using S-transform. Neural networks are then used to compute the classification accuracy of the feature vectors. Certain characteristics of the wavelet feature vectors are apparent from the results. Further in large data sets partitioning is done and similarities of pattern vectors present in different sections are determined. The approach is a general one and can be applied to pattern classification, similarity determination, and knowledge discovery in time varying data patterns occurring in many practical sciences and engineering problems.

Local linear wavelet neural network for breast cancer recognition

Breast cancer is the major cause of cancer deaths in women today and it is the most common type of cancer in women. Many sophisticated algorithm have been proposed for classifying breast cancer data. This paper presents some experiments for classifying breast cancer tumor and proposes the use local linear wavelet neural network for breast cancer recognition by training its parameters using Recursive least square (RLS) approach to improve its performance. The difference of the local linear wavelet network with conventional wavelet neural network (WNN) is that the connection weights between hidden layer and output layer of conventional WNN are replaced by a local linear model. The result quality has been estimated and compared with other experiments. Results on extracted breast cancer data from University of Wisconsin Hospital Madison show that the proposed approach is very robust, effective and gives better classification.

A novel image mining technique for classification of mammograms using hybrid feature selection

The image mining technique deals with the extraction of implicit knowledge and image with data relationship or other patterns not explicitly stored in the images. It is an extension of data mining to image domain. The main objective of this paper is to apply image mining in the domain such as breast mammograms to classify and detect the cancerous tissue. Mammogram image can be classified into normal, benign, and malignant class. Total of 26 features including histogram intensity features and gray-level co-occurrence matrix features are extracted from mammogram images. A hybrid approach of feature selection is proposed, which approximately reduces 75% of the features, and new decision tree is used for classification. The most interesting one is that branch and bound algorithm that is used for feature selection provides the best optimal features and no where it is applied or used for gray-level co-occurrence matrix feature selection from mammogram. Experiments have been taken for a data set of 300 images taken from MIAS of different types with the aim of improving the accuracy by generating minimum number of rules to cover more patterns. The accuracy obtained by this method is approximately 97.7%, which is highly encouraging.

Local linear wavelet neural network based breast tumor classification using firefly algorithm

Breast cancer is the major cause of cancer deaths in women today and it is the most common type of cancer in women. This paper presents some experiments for classifying breast cancer tumor and proposes the use of firefly algorithm (FA) to improve the performance of Local linear wavelet neural network. This work in fact uses FA to optimize the parameters of local linear wavelet neural network. The experiments were conducted on extracted breast cancer data from University of Winconsin Hospital, Madison. The result has been compared with a wide range of classifiers to evaluate its performance. The evaluations show that the proposed approach is very robust, effective and gives better correct classification as compared to other classifiers.

Hybrid approach using KPSO and RLS for RBFNN design for breast cancer detection

Neural networks have been employed in many medical applications including breast cancer classification. Innovation in diagnostic features of tumors may play a central role in development of new treatment methods for earliest stage of breast cancer detection. This study proposes a new hybrid for breast cancer detection by extending the application of a variation of particle swarm optimization called K-particle swarm optimization (KPSO). In this paper, the centers and variances of radial basis functional neural network are initialized by KPSO and then updated using back propagation. The weights are updated using recursive least square instead of back propagation. The results are compared with some recently developed techniques. It is found that the proposed technique provides more accurate result and better classification as compared to some other techniques.

Intelligent system based on local linear wavelet neural network and recursive least square approach for breast cancer classification

A new learning technique for local linear wavelet neural network (LLWNN) is presented in this paper. The difference of the network with conventional wavelet neural network (WNN) is that the connection weights between the hidden layer and output layer of conventional WNN are replaced by a local linear model. A hybrid training algorithm of Error Back propagation and Recursive Least Square (RLS) is introduced for training the parameters of LLWNN. The variance and centers of LLWNN are updated using back propagation and weights are updated using Recursive Least Square (RLS). Results on extracted breast cancer data from University of Wisconsin Hospital Madison show that the proposed approach is very robust, effective and gives better classification.

Retraction Note to: Mass classification method in mammograms using correlated association rule mining

In this paper, we present an efficient computer-aided mass classification method in digitized mammograms using Association rule mining, which performs benign–malignant classification on region of interest that contains mass. One of the major mammographic characteristics for mass classification is texture. Association rule mining (ARM) exploits this important factor to classify the mass into benign or malignant. The statistical textural features used in characterizing the masses are mean, standard deviation, entropy, skewness, kurtosis and uniformity. The main aim of the method is to increase the effectiveness and efficiency of the classification process in an objective manner to reduce the numbers of false-positive of malignancies. Correlated association rule mining was proposed for classifying the marked regions into benign and malignant and 98.6% sensitivity and 97.4% specificity is achieved that is very much promising compare to the radiologist’s sensitivity 75%.

A self-adaptive fuzzy-based optimised functional link artificial neural network model for financial time series prediction

In recent years, new data mining and machine learning techniques have been developed and applied to various fields of science. Out of these recently developed techniques few offer online support and are able to adapt to large and complex financial dataset. Therefore, the present research adopts Functional Link Artificial Neural Network (FLANN) model for predicting the closing price of three companies namely Yahoo Inc, Nokia and Bank of America. The FLANN model used is trained by fuzzy after normalisation of the data and closing price is forecasted for one day and one week ahead. The prediction result is compared with the parameters of the FLANN model trained by Genetic Algorithm (GA) and Particle Swarm Optimisation (PSO). The proposed training method provides better accuracy and takes less time as compared to training the FLANN model using PSO or GA. The proposed approach has also been compared with a linear dataset for validation. The FLANN-fuzzy approach is seen to provide better results in predicting financial distress.

A harmony search-based artificial neural network for stock market prediction

For financial time series, the generation of error bars on the point prediction is important in order to estimate the corresponding risk. In recent years, artificial intelligence optimisation techniques have been used to make time series approaches more systematic and improve forecasting performance. The harmony search learning methodology, already successfully applied for training of multilayer perceptrons, is applied to Functional Link Artificial Neural Network (FLANN) in order to infer non-linear models for predicting a time series and the related volatility. The proposed method is implemented and the results are compared with FLANN model trained by back propagation and differential evolution. The proposed training method shows that FLANN-harmony search provides better forecasting/prediction as compared to training the FLANN model using back propagation or differential evolution.