Qualification: 
Ph.D, MSc, M.Tech
n_madhusudanarao@cb.amrita.edu

Dr. Madhusudana Rao Nalluri joined the Department of Mathematics, School of Engineering, Amrita Vishwa Vidyapeetham, Coimbatore campus, in the year 2019.
He is currently an Assistant Professor (Selection Grade) with the Department of Computer Science and Engineering.

Earlier, he worked as Assistant Professor-III in the School of Computing of Sastra University for 9 years. Before joining Sastra, he worked as Senior Lecturer in the Computer Science and Engineering department of the National Institute of Science and Technology during the academic year 2009-10.

He received his M. Sc. degree in Pure Mathematics from Andhra University, India. He has obtained 84th rank in Gate-2007. He received his M.Tech. degree in Scientific Computing from Birla Institute of Technology India. He earned his Ph. D. degree from Sastra University, in February 2019. His Ph. D. thesis title is “Development of Hybrid Computational Intelligence Techniques for Multi-Objective Optimization of Classification Models using Enhanced Evolutionary Algorithms”. He has authored around 17 technical papers in reputed conferences and journals indexed in Scopus and SCI. His cumulative impact factor of SCI-indexed publications is 12.825.

Education

  • 2019: Ph. D.(Computer Science and Engineering)
    Sastra University
  • 2009: M. Tech. (Scientific Computing)
    Birla Institute of Technology
  • 2005: M. Sc. (Pure Mathematics)
  • Andhra University

Publications

Publication Type: Journal Article

Year of Publication Title

2020

Madhusudana Rao Nalluri, Kannan, K., Gao, X. - Z., and Roy, D. Sinha, “Multiobjective hybrid monarch butterfly optimization for imbalanced disease classification problem”, International Journal of Machine Learning and Cybernetics volume , vol. 11, no. 7, pp. 1423 - 1451, 2020.[Abstract]


Datasets obtained from the real world are far from balanced, particularly for disease datasets, since such datasets are usually highly skewed having a few minority classes apart from one or more prominent majority classes. In this research, we put forward the novel hybrid architecture to handle imbalanced binary disease datasets that arrives upon the efficient combination of Support vector machine (SVM) classifier’s sensitive parameter values for improved performance of SVM by means of an Evolutionary algorithm (EA), namely monarch butterfly optimization (MBO). In this paper, MBO is used to enumerate three objectives, namely prediction accuracy (PAC), sensitivity (SEN), specificity (SPE). Additionally, we propose a Totally uni-modular matrix (TUM) and limit points based non-dominated solutions selection for deciding local and global search and to generate an efficient initial population respectively. Since these two greatly affect the performance of EAs, the performance of the proposed hybrid architecture is tested on 18 disease datasets having binary class labels and the results obtained demonstrate improvements using the proposed method. For the majority of the datasets, either 100% sensitivity and/or specificity were attained. Moreover, pertinent statistical tests were carried out to ascertain the performances obtained.

More »»

2019

Madhusudana Rao Nalluri, Kannan K., Xiao-Zhi Gao, and Diptendu Sinha Roy, “An efficient hybrid meta-heuristic approach for cell formation problem”, Soft Computing, vol. 23, no. 19, pp. 9189 - 9213, 2019.[Abstract]


The cellular manufacturing technology, an application of group technology in manufacturing, has been a widely studied combinatorial optimization problem where the entire production system is divided into many cells and part families. In this paper, a novel clonal selection algorithm (CSA) that uses a new affinity function and part assignment heuristic for solving a multi-objective cell formation problem is studied. The proposed CSA has been hybridized with genetic algorithm for generating feasible cell sequences that fulfill both mutual exhaustivity and exclusion properties of machine cells prior to the initial population generation. Additionally, a new part assignment heuristic function that maps parts to machine cells and a novel basic affinity function have been built into the proposed CSA so that it can act as the utility function to solve the multi-objective cell formation problem. This hybrid CSA (HCSA) has been presented and computational results have been obtained for the proposed scheme with a set of 52 benchmark instances collected from literature. The results presented herein demonstrate that overall proposed HCSA is much more promising in comparison with existing approaches available in recent literatures. Extensive statistical and convergence tests have been carried out to ratify the superiority of the proposed HCSA. The improvements can be attributed to the collaborative interactions in the CSA mechanism, the proposed hybridization for initial population generation and so forth.

More »»

2019

Saru Meena Ramu, Muthaiah Rajappa, Kannan Krithivasan, and Madhusudana Rao Nalluri, “A novel fast medical image segmentation scheme for anatomical scans”, Multimedia Tools and Applications, vol. 78, no. 15, pp. 21391–21422, 2019.[Abstract]


Medical image is the visual representation of anatomy or physiology of internal structures of the body and it is useful for clinical analysis and medical intervention. Modern medical imaging devices provide an excellent view of anatomy and physiology of internal structures of the body non-invasively. However, the usage of computers to measure, examine and determine the state of internal structures of the body with accuracy and efficiency is limited. Automated medical image segmentation techniques have wide range of utility in diagnosis, treatment planning and computer integrated surgery. These automated medical image segmentation techniques could also be used as an assisting tool to radiologists by saving their time in selecting, measuring and classifying various findings. However, automated medical image segmentation is challenging because the quality of the image is low due to the presence of noise, artefacts, partial volume effects etc., low contrast between different structures in an image and intensity variations within a region itself. This research paper focuses on fastening a region based deformable model called Chan-Vese model through various first order optimization techniques. Chan - Vese model can perform segmentation effectively even in low quality images but the limitation of Chan-Vese model is that convergence towards optimal solution is slow. The objective of this work is to fasten the convergence of Chan-Vese model towards optimal solution by using various first order optimization schemes. Chan-Vese model with proposed optimization techniques is tested with X-ray, CT and MRI images of different organs. Comparative study between traditional optimization technique used in Chan-Vese model and proposed optimization techniques has been carried out. From the comparative study, it is found that Chan-Vese model with proposed optimization schemes is efficient in terms of speedy delineation with less number of iterations and processing time. Therefore, this fastened Chan-Vese model is better suited algorithm for fast image segmentation needs such as tracking of region of interest in subsequent frames in a video.

More »»

2018

Madhusudana Rao Nalluri, Krithivasan Kannan, Xiao-Zhi Gao, and Diptendu Sinha Roy, “Novel classifiers for intelligent disease diagnosis with multi-objective parameter evolution”, Computers & Electrical Engineering, vol. 67, pp. 483 - 496, 2018.[Abstract]


In this research, intelligent classifiers for disease diagnosis are designed that use classifier parameters, such as cost, tolerance, gamma and epsilon, with multi-objective evolutionary algorithms. The multiple objective functions are prediction accuracy, sensitivity and specificity. This paper employs a Sequential Minimal Optimization (SMO), a variant of the classical Support Vector Machine (SVM), as the base classifier in conjunction with three popular evolutionary algorithms (EA), namely, Elephant Herding Optimization (EHO), Multi-objective Evolutionary Algorithm Based on Decomposition (MOEA/D) and Non-dominated Sorting Genetic Algorithm-II (NSGA-II), for parameter evolution. A new cuboids based initial population generation mechanism was also introduced to hybridize EHO, called CEHO. The performance of CEHO is compared with the other three EAs (EHO, MOEA/D and NSGA-II) over 17 medical engineering datasets, and pertinent statistical tests were conducted to substantiate their performances. The results demonstrate that the proposed CEHO exhibit better to competitive results across all datasets.

More »»

2017

Madhusudana Rao Nalluri, Kannan K., Manisha M, and Diptendu Sinha Roy, “Hybrid Disease Diagnosis Using Multiobjective Optimization with Evolutionary Parameter Optimization.”, Journal of Healthcare Engineering, vol. 2017, p. 5907264, 2017.[Abstract]


With the widespread adoption of e-Healthcare and telemedicine applications, accurate, intelligent disease diagnosis systems have been profoundly coveted. In recent years, numerous individual machine learning-based classifiers have been proposed and tested, and the fact that a single classifier cannot effectively classify and diagnose all diseases has been almost accorded with. This has seen a number of recent research attempts to arrive at a consensus using ensemble classification techniques. In this paper, a hybrid system is proposed to diagnose ailments using optimizing individual classifier parameters for two classifier techniques, namely, support vector machine (SVM) and multilayer perceptron (MLP) technique. We employ three recent evolutionary algorithms to optimize the parameters of the classifiers above, leading to six alternative hybrid disease diagnosis systems, also referred to as hybrid intelligent systems (HISs). Multiple objectives, namely, prediction accuracy, sensitivity, and specificity, have been considered to assess the efficacy of the proposed hybrid systems with existing ones. The proposed model is evaluated on 11 benchmark datasets, and the obtained results demonstrate that our proposed hybrid diagnosis systems perform better in terms of disease prediction accuracy, sensitivity, and specificity. Pertinent statistical tests were carried out to substantiate the efficacy of the obtained results.

More »»

Publication Type: Conference Proceedings

Year of Publication Title

2020

S. Divya, Eranki L. N. Kiran, Madhusudana Rao Nalluri, and Pujitha Vemulapati, “Prediction of Gene Selection Features Using Improved Multi-objective Spotted Hyena Optimization Algorithm”, Data Communication and Networks. Springer Singapore, Singapore, pp. 59-67, 2020.[Abstract]


Microarray data analysis is one of the main researchDivya, S. areas in the medical research. The Microarray is a dataset which consists of different geneKiran, Eranki L. N. expressions from which most of the features areRao, Madhu Sudana redundant genes and reducing the classifier accuracy. Finding a minimal subset of features from large geneVemulapati, Pujitha expression is a challenging task where removing redundant feature but the important feature will not be missed. Many optimization techniques are introduced by the researchers to find a minimal subset of features but it does not provide a feasible solution. In this paper, the RWeka package, which provides an interface of Weka tool functionality to R is used to order the features using select attribute function in Weka. By using those ordered features, a minimal subset of features is selected using SVM classifier with maximum prediction accuracy in the dataset. Obtained minimal subset of features is given as input to the Multi-Objective Spotted Hyena Optimizer algorithm which is driven by the ensemble of SVM classifier by updating the search agents with objective function with an intension to improve the classification accuracy. The proposed method has experimented with seven publicly available microarray datasets such as CNS, colon, leukemia, lymphoma, lung, MLL, and SRBCT, which shows that the proposed methodology gives the high accuracy than all other existing techniques in terms of feature selection and prediction accuracy.

More »»

2020

M. Rajasekhar Reddy, B. Nithish Kumar, Madhusudana Rao Nalluri, and B. Karthikeyan, “A New Approach for Bias–Variance Analysis Using Regularized Linear Regression”, Advances in Bioinformatics, Multimedia, and Electronics Circuits and Signals. Springer Singapore, Singapore, pp. 35-46, 2020.[Abstract]


Bias–variance is one of the tools to learn the performance of any machine learning algorithm. Various bias–variance models can be observed using regularized linear regression. In this paper, we implement regression on a sample dataset along with some optimization technique. Here, we use a technique called polynomial linear regression on the same dataset to increase the fit and the results will be normalized. Earlier studies show that though the polynomial linear regression is applied, the plot is very complex and it will drop-off at extremes at most of the situations. Regularization is one of the techniques used to optimize and narrower the gaps. These properties vary from one dataset to another and we find an optimistic value of parameter lambda. The error of overfitting from the plotted curve is significantly reduced. Again it is plotted between error obtained and the values of lambda. The main goal of this paper is to avoid the overfitting problem. Cross-validation set is taken and helped in estimation of error and deciding the parameters work best in the model.

More »»

2017

Madhusudana Rao Nalluri, T. SaiSujana, K. H. Reddy, and V. Swaminathan, “An efficient feature selection using artificial fish swarm optimization and svm classifier”, 2017 International Conference on Networks Advances in Computational Technologies (NetACT). 2017.[Abstract]


The medical datasets have many features if the features have a tendency of mutation then the risk of disease increases which makes difficult to provide a diagnosis of disease. In the dataset, every feature is a contributor for prediction accuracy, the selection of significant features from the dataset is a challenging task. The feature selection technique based on metaheuristic algorithms is used for the selection of significant data. The metaheuristic algorithm inspired by the fish behavior under water is artificial fish swarm optimization (AFSO) is proposed in this paper. The wrapper approach of AFSO with support vector machine (SVM) is used for finding the feature subset with a minimum number of features. The proposed approach is tested on nine different datasets having binary and multiple imbalanced classes and correlated with other metaheuristic algorithms. The results show that the proposed approach is providing high classification accuracy with features subset having fewer features.

More »»

2017

T. S. Sujana, Madhusudana Rao Nalluri, and R. S. Reddy, “An efficient feature selection using parallel cuckoo search and naïve Bayes classifier”, 2017 International Conference on Networks Advances in Computational Technologies (NetACT). 2017.[Abstract]


In real world, the datasets are having varying dimensions which incorporates noisy, irrelevant and redundant data which is hard to analyze. Feature selection is a preprocessing step used for selecting the significant information. The selection of optimal feature subset is an optimization problem which has been solved by several versions of metaheuristic algorithms. The metaheuristic optimization algorithm based on the behavior of cuckoo birds is adapted to build the parallel cuckoo search optimization (PCSO) algorithm. The wrapper approach of parallel cuckoo search with Naive Bayes (PCSNB) is developed by combining the power of exploration of PCSO with the speed of Naïve Bayes (NB) classifier for finding feature subset that maximizes the accuracy. The proposed approach is tested on seven different datasets which are having balanced and imbalanced classes and contrasted with other metaheuristic algorithms. The results are showing higher prediction accuracy than other algorithms and selects the feature subset with less features.

More »»

2017

S. D. Suganya, M. R. Reddy, and Madhusudana Rao Nalluri, “Increasing the quality of reconstructed image through hybrid compression technique”, 2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS). 2017.[Abstract]


A novel image compression algorithm based on the combination of three different techniques is presented. First, the image is compressed using PCA and then DCT is applied to the reconstructed image acquired from PCA technique. The reconstructed image from the block based DCT method is further compressed through the DWT based Set Partitioning In Hierarchical Trees (SPIHT) compression technique. At last, output reconstructed image will be evaluated with variety of quality metrics. The proposed novel technique is evaluated with the various quality metrics namely PSNR, SC, SNR. The PSNR value of the proposed technique is high, when compared with other existing techniques. The experiment reveals that the proposed compression method improves the quality of the reconstructed image.

More »»

2013

Madhusudana Rao Nalluri, M. Aruna, and S. Bhuvaneswari, “Meta Heuristic Approaches for Circular Open Dimension Problem”, Swarm, Evolutionary, and Memetic Computing. Springer International Publishing, Cham, pp. 44-54, 2013.[Abstract]


This paper discusses the circular open dimension problem (CODP), where set of circles of different radii has to be packed into a rectangular strip of predetermined width and variable length. The circle packing problem is one of the variant of cutting and packing problems. We propose four different nature inspired Meta heuristic algorithms for solving this problem. These algorithms are proved to be the best in finding local solutions. The algorithms are based on food foraging process and breeding behavior of some biological species such as bat, bee, firefly and cuckoo. Circle packing problem is one of the NP hard problems. It is very difficult to solve NP hard problems exactly, so the proposed approaches tries to give approximate solution within the stipulated time. The standard benchmark instances are used for comparison, and it is proved that firefly is giving the best solution.

More »»

2012

Madhusudana Rao Nalluri, D. S. Roy, P. R. Parro, and D. K. Mohanta, “An ant colony metaheuristic approach for optimal reliability assessment of software systems incorporating redundancy”, 2012 World Congress on Information and Communication Technologies. 2012.[Abstract]


With the all pervasive presence of computers to all aspects of life, software reliability assessment is assuming a position of utmost importance. Moreover many commercial and governmental software systems require high mission reliability requiring both hardware and the software to be very reliable. Software reliability is acknowledged to perk up with the amount of testing efforts invested, which in turn reduces the cost of software development and in turn system cost. The scale of redundancy employed affects reliability favorably, while increasing the cost of software design and development. This paper employs an ant colony meta-heuristic optimization method to solve the redundancy allocation problem (RAP) for software systems. Herein, an ant colony optimization algorithm for the software RAP (SRAP) is devised and tested on a computer relay software that is employed for fault handling in power system transmission lines and the results presented validates the efficacy of the approach.

More »»

2011

V. M. K. Prasad Goura, Madhusudana Rao Nalluri, and M. Rajasekhar Reddy, “A dynamic clustering technique using minimumspanning tree”, 2nd International Conference on Biotechnology and Food Science IPCBEE, vol. 7. IACSIT Press, Singapore, 2011.[Abstract]


Clustering technique is one of the most important and basic tool for data mining. In this paper, we present a clustering algorithm that is inspired by minimum spanning tree. Given the minimum spanning tree over a data set, selects or rejects the edges of the MST in process of forming the clusters, depending on the threshold value. The Algorithm is invoked repeatedly until all the clusters are fully formed. We present experimental results of our algorithm on some synthetic data sets as well as real world data sets.

More »»

2011

Madhusudana Rao Nalluri, V. M. K. P. Goura, D. S. Roy, and D. K. Mohanta, “A binary integer programming solution for optimal reliability of computer relaying software incorporating redundancy”, 2011 IEEE Recent Advances in Intelligent Computational Systems. 2011.[Abstract]


Software reliability assessment is an area of prime importance of today, more so since computers have pervaded to practically all aspects of our lives. Moreover many software based commercial and governmental systems require very high mission reliability. To meet this, both the hardware as well as the software has to be extremely reliable. Software reliability is known to improve with the amount of testing efforts invested, which in turn leverages the cost of software development and in turn system cost. Constructing systems employing redundant inferior units that are relatively cheap is a well known alternative. Using the N-version programming paradigm, the same can be applied to software, but the cost factor becomes a major issue. The degree of redundancy employed affects reliability favorably, while increasing the cost of software design and development. In this paper, an approach to enumerate the optimal redundancy for software systems has been explored and a binary integer programming solution for the same has been presented. The methodology has been applied to the case study of a computer relay software that is employed for fault handling in power system transmission lines.

More »»

2011

Madhusudana Rao Nalluri, Diptendu Sinha Roy, and Dusmanta K. Mohanta, “Application of NSGA - II to Power System Topology Based Multiple Contingency Scrutiny for Risk Analysis”, SEMCCO 2011: Swarm, Evolutionary, and Memetic Computing. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 706-713, 2011.[Abstract]


The Incorporation of deregulation and increase in renewable sources of generation has shifted the nature of existing power systems to a more geographically distributed system. This had led to significant challenges towards on-line monitoring and control. Contingency set identification is an essential step in monitoring the power system security level. Multiple contingency analysis forms the basis of security issues, particularly of large, interconnected power systems. The difficulty of multiple contingency selections for on-line security analysis lies in its inherent combinatorial nature. In this paper, an approach for identification of power system vulnerability to avoid catastrophic failures is put forward, as a multi objective optimization problem that partitions its topology graph, accounts for maximizing the imbalance between generation and load in each island and at the same time minimizes the number of lines cut to realize the partitions. The Nondominated Sorted Genetic Algorithm, version II (NSGA II) has been applied to obtain the optimal solutions and the methodology involved has been applied to an IEEE 30 bus test system and results are presented.

More »»