Eduation
Ph.D. in Computer Science & Engineering
University College of Engineering, Jawaharlal Nehru Technological University, JNTU Kakinada (in Collaboration with Tennessee Technological University, USA)
Thesis Title: Novel Graph-Based Optimization Approaches for Finding Interesting Substructures in Heterogeneous Networks
My doctoral research, titled "Novel Graph-Based Approaches for Finding Interesting Substructures in Heterogeneous Networks," conducted at Jawaharlal Nehru Technological University, Kakinada, focused on advancing the field of graph-based knowledge discovery. I made significant contributions to various areas:
Firstly, I introduced the Minimum Description Length (MDL) metric to evaluate substructures in graph datasets. This included applying MDL encoding to diverse graph datasets, solving the graph matching problem using MDL, and conducting statistical analyses to understand the variation of MDL values with different graph properties. I addressed the challenge of mining anomalous substructures by proposing a novel graph-based anomaly detection approach. This approach utilized background knowledge rules, incorporating MDL and size metrics to aid in the discovery of anomalous substructures. Furthermore, I demonstrated the effectiveness of anomaly detection in improving classification accuracy and selecting discriminating substructures through empirical evaluations on both synthetic and real-world datasets, including resumes, chemical compounds, and golf domains.
Moreover, I developed novel domain-independent rules for mining interesting patterns. These rules, such as "netting" and "slackness," were designed to report external networks and mine acyclic structures in a graph, respectively. Additionally, I proposed CRADLE-MDL and CRADLE-Size methods that leverage MDL and size metrics to aid in discovering interesting substructures, contributing to the advancement of graph mining techniques.
Lastly, my research extended to mining patterns from educational data using conceptual graphs. Here, I formulated the problem of common skill-set extraction from resumes and proposed a MapReduce algorithm to convert preprocessed data into conceptual graphs. Applying SUBDUE, a graph mining algorithm, to the skill-sets analytics domain, I identified common skills and evaluated the performance of the proposed approach through experimental results. This sequential flow of novel algorithms, rules, and applications demonstrates the breadth and significance of my contributions to the field of graph-based knowledge discovery.
Novel Graph-Based Approaches for Real-Time Problem Solving
Anomaly Detection: Introducing a novel graph-based anomaly detection approach, the research aids in identifying anomalous patterns in datasets. This has applications in diverse fields such as healthcare, networks, finance, and insurance, where detecting anomalies is crucial for ensuring data integrity and security.
Skill-Set Analytics: The application of SUBDUE, a graph mining algorithm, to the skill-set analytics domain enables the identification of common skills from resumes. This addresses the real-time challenge faced by recruiters and employers in efficiently extracting relevant information from large volumes of resumes.
Educational Data Mining: The approach to mining educational data using conceptual graphs contributes to the field of educational data mining. It helps in extracting valuable insights from educational datasets, facilitating personalized learning, curriculum development, and educational policy-making.
Improving Classification Accuracy: Through the incorporation of background knowledge rules and metrics like MDL and size, the research enhances classification accuracy. This is beneficial in scenarios where precise classification is essential, such as in healthcare diagnosis, fraud detection, or customer segmentation.
Network Analysis: The application of the methods in graph-based knowledge discovery aids in network analysis, where understanding complex relationships and structures is vital. This has implications for social network analysis, cybersecurity, and infrastructure optimization.
M.Tech. in Computer Science & Engineering
University College of Sciences, Acharya Nagarjuna University, ANU, Guntur
B.Tech. in Computer Science & Engineering
Jawaharlal Nehru Technological University, JNTU Kakinada
Awards
Bharat Education Excellence Award 2024
Recognized for outstanding contributions to the educational society with the Bharat Education Excellence Award 2024.
The selection process for the BEEA was highly competitive, with an overwhelming response of over 8,000 nominations received. After meticulous consideration, the selection committee carefully reviewed each nomination based on criteria such as impact, innovation, and leadership in the field of education.
Following this initial review, the committee conducted a thorough evaluation of the top candidates, considering factors such as the quality of contributions, impact on the educational community, and alignment with the values of the BEEA. Through this rigorous process, the committee identified the top 300 profiles that best exemplified excellence in education. I am honored to share that I am one among these distinguished 300 profiles.
Best Faculty Award-2010
Honored with the Best Faculty Award in 2010 from QIS Institute of Technology, Ongole.
I am honored to have received the Best Faculty Award on September 5th, 2010, from QIS Institute of Technology, Ongole, on the occasion of Teachers' Day. This prestigious award recognizes excellence in teaching and contributions to the academic community.