Graph coloring, Machine Learning, Data Science and Bigdata Analytics
12
Scopus Publications
Scopus Publications
L(2, 1)-coloring and its related problems for Mycielskians of certain classes of graphs Srinivasa Rao Kola, Balakrishna Gudla, P. K. Niranjan Discrete Mathematics Algorithms and Applications, 2025 In an [Formula: see text]-coloring [Formula: see text] of [Formula: see text] with span [Formula: see text], [Formula: see text] is a hole if there is no vertex [Formula: see text] in [Formula: see text] such that [Formula: see text]. An [Formula: see text]-coloring with span [Formula: see text] is said to be a no-hole coloring if it uses all the colors from [Formula: see text]. An [Formula: see text]-coloring of a graph [Formula: see text] is irreducible if color of no vertex can be decreased and yield another [Formula: see text]-coloring of [Formula: see text]. A graph [Formula: see text] is inh-colorable if there exists an irreducible no-hole coloring of [Formula: see text]. The inh-span of a graph [Formula: see text], denoted by [Formula: see text] is the smallest number [Formula: see text] such that there is an irreducible no-hole [Formula: see text]-coloring of [Formula: see text] with span [Formula: see text]. The maximum number of holes in [Formula: see text] is the maximum number of holes over all irreducible span colorings of [Formula: see text]. In this paper, we give an algorithm to define an [Formula: see text]-coloring for the Mycielskian [Formula: see text] of an arbitrary graph [Formula: see text]. We prove that the algorithm gives span colorings for Mycielskians of path [Formula: see text] ([Formula: see text]), cycle [Formula: see text] ([Formula: see text]), some classes of trees and concatenation of an arbitrary graph with trees. Also, we show that the Mycielskians for the above classes of graphs are inh-colorable and their inh-span is same as the [Formula: see text]-number. Further, we determine the maximum number of holes over all irreducible span colorings for Mycielskians of path [Formula: see text] ([Formula: see text]), cycle [Formula: see text] ([Formula: see text]), some classes of trees and concatenation of an arbitrary graph with trees.
Optimizing Network Security with SEB-ID: Hybrid Ensemble Learning Framework for Accurate and Scalable Anomaly Detection Sayyada Mubeen, Balakrishna Gudla Iet Conference Proceedings, 2025 The growing complexity of cyberattacks has made network security a major worry for businesses all around the world. Finding possible network problems and malicious activity requires the application of anomaly detection. These anomalies have been discovered using conventional machine learning (ML) techniques, with models like Random Forest and Convolutional Neural Networks (CNN) showing promise. Yet there are problems with the current approaches, such as poor feature selection, insufficient data pre-processing, and a dearth of efficient ensemble learning strategies. By putting forth a Smart Ensemble-Based Intrusion Detection (SEB-ID) architecture for network anomaly detection, this research aims to overcome these constraints. To increase detection accuracy and robustness, several machine-learning classifiers are combined in the proposed model employing machine-learning classifiers using ensemble-learning approaches, including Decision Tree, Random Forest, ExtraTrees, and XGBoost. The Synthetic Minority Oversampling Technique (SMOTE) is used to balance the dataset, handle missing values, and normalize it as part of a thorough pre-processing phase. To improve model performance, hyperparameter adjustment is proposed via Bayesian optimization. Important performance metrics such as precision, accuracy, F1-score and recall to evaluate SEB-ID model's efficiency, guaranting a strong and dependable detection system.
Enhancing K-means Clustering Accuracy based on Robust and Abs scaling Maradana Durga Venkata Prasad, Srikanth T, Balakrishna Gudla Iet Conference Proceedings, 2025 We discuss a new narrative custom scaling method employed improving to improve the outcome of K-Means Clustering in the current article. The new method, referred to as custom scaling, is defined as X′= X−Median(X) / max(X), a combination of the robustness of median-based scaling and the normalization provided by max-based scaling. This approach effectively reduces the influence of outliers and ensures that feature values are appropriately normalized. The comparison of K-means clustering outcome is often made with traditional scaling methods (standard scaling and min-max scaling) and our custom scaling technique across several datasets. Experiments reveal the facts on how the proposed custom scaling method significantly improves the accuracy of clustering, especially for datasets with outliers.
ADAPTIVE ROUTING STRATEGIES FOR ENERGY CONSERVATION USING POWER SELF-OPTIMIZATION TECHNIQUE IN WIRELESS SENSOR NETWORKS G. Mohan Ram, Balakrishna Gudla, Meeravali Shaikh, S. Naveen Kumar Iet Conference Proceedings, 2025 Wireless Sensor Networks (WSNs) have received considerable focus from few years because of their diverse set of domains including military surveillance, environmental monitoring, and healthcare. These networks offer efficient data collection and transmission capabilities. However, a major challenge in WSNs is energy consumption, particularly during data transmission. The network's effectiveness largely relies on the availability of functional nodes. If inactive or dead nodes exist along a designated transmission path, communication delays and increased energy consumption may occur. Additionally, transmitting data at high power levels can cause interference, leading to inefficient transmission and power wastage.
Classification of Human Ear Shape with Innovative RF and SVM Machine Learning Algorithms and Related Issues M.S Saravanan, Balakrishna Gudla, Y.R Sampath Kumar, K.Jane Nithya, M.R. Arun 2024 2nd International Conference on Computer Communication and Control Ic4 2024, 2024 The main objective of this paper is to improve the accuracy for especial features of ear detection with ear shape images using learning algorithms. Materials and Methods: This paper’s dataset consists of 1416 ear shape images in order to categorise of ear detection. The data are labelled as “ear shape”, “ear size”, “left ear”, “right ear”, image classification from the images into these types, 20 number of images have been used for Random Forest (RF) algorithm taken as case as one and is compared with Support Vector Machine (SVM) algorithm taken as case two with ear shape repository images and it collected in Kaggle website and the parameters are fixed with 95% Confidence Interval and 0.05% significance value. Results: This research study uses two cases of algorithms in this Random Forest (RF) algorithm has achieved an improved accuracy of 98%, compared to the SVM algorithm of 89% and a significant value of 0.028 with a 95% confidence interval. Conclusion: This study discovered that the Random Forest (RF) algorithm predicts ear detection significantly better than the SVM algorithm.
Exploring Next-Generation Internet with Decentralized Web Architectures and Web Technologies W. Sarada, Pramod Kumar, G.Sunitha Rekha, B. Aishwarya, R.Niranjana, Balakrishna Gudla 3rd International Conference on Advances in Computing Communication and Materials Icaccm 2024, 2024 The advances in the architecture of the new web and propelled by Web3 innovations are the major playground where the Internet is moving at a pace faster than we could imagine. It is towards that direction that this research seeks to examine the effectiveness of decentralised systems with emphasis on how Blockchains, Smart Contracts and Peer to Peer systems can revolutionalise the internet by offering more security, transparency and user ownership. Web3 technologies pin the inconvenience of Web2 as being too centralized, lacking in privacy, and problematic in terms of data ownership; Web3 can champion decentralized applications [or dApps] and smart contracts. Taking the key blockchain camps including Ethereum, Solana, and Polkadot into consideration, the study measures their effectiveness, operation capability, and security features. Success and failure rates of the decentralized architecture concept in response to problems affecting different sectors such as finance, healthcare, and digital ownership are measured by simulations as well as the case studies. Particular emphasis is created on such issues as scalability, legal requirements, and user acceptance. Application blockchains are audited by means of blockchain analytics tools and smart contract testing frameworks to identify the strengths and weaknesses of DApps in order to improve them. This research also points towards the issues that require focusing on the improvement of governance models, UX interfaces, and security frameworks in the Web3 sphere. By assessing the existing decentralized platforms and perform performance study, the research identifies the drawbacks and challenges of the blockchain platform and suggests probable solutions to tackle them. Thus, the study results indicate that decentralized web architectures as such have a rather high potential in the future, although the issues of scalability and usability will become the main factors to define their success.
Interference Management Coverage and Capacity Optimization in Hetnets R. S. Arunkumar, V. Mohamed Keeran, K. Pavithra, Balakrishna Gudla, A. Raajya Vardhini, Nandhini T J 3rd International Conference on Advances in Computing Communication and Materials Icaccm 2024, 2024 The ever-increasing demand for high data rates, ubiquitous connectivity, and enhanced Quality of Service (QoS) in modern wireless communication systems has given rise to the emergence of a potentially fruitful solution in the form of heterogeneous networks, often known as HetNets. HetNets provide improved coverage, capacity, and overall network performance by combining classic macrocells with a variety of small cell types, such as microcells and picocellss, in addition to the traditional macrocells. However, the occurrence of multiple cell types in close proximity presents difficulties in the management of interference, the optimisation of coverage, and the distribution of available capacity. The primary objective of this research is to devise innovative interference management strategies, intelligent resource allocation algorithms, and adaptive network topologies with the goal of resolving the issues that arise in heterogeneous networks (HetNets). We plan to improve network performance and the overall user experience by making use of cutting-edge technology such as artificial intelligence and machine learning to reduce interference problems, improve coverage patterns, and more effectively manage resources. The importance of this study lies in the fact that it reveals the full potential of HetNets, thereby making it possible for those networks to fulfil the ever-increasing requirements of data-intensive applications and services. HetNets have the ability to sustain variable user densities and usage patterns, as well as deliver smooth connectivity and reliable services. This is made possible through optimising interference control, coverage, and capacity. The findings of this study make a contribution to the construction of a wireless communication infrastructure that is more sustainable and future-proof, and that can accommodate the varied requirements of mobile users both now and in the future. In the end, the goal of our effort is to encourage the development of HetNets in a way that is both efficient and dependable, so enabling wireless networks to fulfil the ever-shifting requirements of the digital era
Some classes of trees with maximum number of holes two Srinivasa Rao Kola, Balakrishna Gudla, Niranjan P.K. Akce International Journal of Graphs and Combinatorics, 2020 An L2,1-coloring of a simple connected graph G is an assignment of non-negative integers to the vertices of G such that adjacent vertices color difference is at least two, and vertices that are at distance two from each other get different colors. The maximum color assigned in an L2,1-coloring is called span of that coloring. The span of a graph G denoted by λG is the smallest span taken over all L2,1-colorings of G. A hole is an unused color within the range of colors used by the coloring. An L2,1-coloring f is said to be irreducible if no other L2,1-coloring can be produced by decreasing a color of f. The maximum number of holes of a graph G, denoted by HλG, is the maximum number of holes taken over all irreducible L2,1-colorings with span λG. Laskar and Eyabi (Christpher, 2009) conjectured that if T is a tree, then HλT=2 if and only if T=Pn, n>4. We show that this conjecture does not hold by providing a counterexample. Also, we give some classes of trees with maximum number of holes two.
Infinitely Many Trees with Maximum Number of Holes Zero, One, and Two Srinivasa Rao Kola, Balakrishna Gudla, P. K. Niranjan Journal of Applied Mathematics, 2018 An L(2,1)-coloring of a simple connected graph G is an assignment f of nonnegative integers to the vertices of G such that fu-fv⩾2 if d(u,v)=1 and fu-fv⩾1 if d(u,v)=2 for all u,v∈V(G), where d(u,v) denotes the distance between u and v in G. The span of f is the maximum color assigned by f. The span of a graph G, denoted by λ(G), is the minimum of span over all L(2,1)-colorings on G. An L(2,1)-coloring of G with span λ(G) is called a span coloring of G. An L(2,1)-coloring f is said to be irreducible if there exists no L(2,1)-coloring g such that g(u)⩽f(u) for all u∈V(G) and g(v)<f(v) for some v∈V(G). If f is an L(2,1)-coloring with span k, then h∈0,1,2,…,k is a hole if there is no v∈V(G) such that f(v)=h. The maximum number of holes over all irreducible span colorings of G is denoted by Hλ(G). A tree T with maximum degree Δ having span Δ+1 is referred to as Type-I tree; otherwise it is Type-II. In this paper, we give a method to construct infinitely many trees with at least one hole from a one-hole tree and infinitely many two-hole trees from a two-hole tree. Also, using the method, we construct infinitely many Type-II trees with maximum number of holes one and two. Further, we give a sufficient condition for a Type-II tree with maximum number of holes zero.
Local binary patterns for gender classification Balakrishna Gudla, Srinivasa Rao Chalamala, Santosh Kumar Jami Proceedings Aims 2015 3rd International Conference on Artificial Intelligence Modelling and Simulation, 2016
