Qualification: 
MCA
deepa@asas.kh.amrita.edu

G. Deepa currently serves as Assistant Professor in the Department of Computer Science and I.T., School of Arts & Sciences, Kochi. She has 9 years experience in teaching. 

Publications

Publication Type: Journal Article

Year of Publication Title

2019

Sruthi S Menon, Mary Saana N J, and G. Deepa, “Image Forgery Detection using Hash Functions”, International Journal of Recent Technology and Engineering (IJRTE) , vol. 8, no. 1, 2019.[Abstract]


Digital images are widely spread in today’s world and morphing of these images are also increased. Morphing of the images is the process of changing original image into another image using different tools. In social media the invasion of these morphed images are rapidly increasing and traditionally, the tampered images were found by the pixel comparison method. This way of detection leads to complexity and space consumption. pHash is used in this system as hashing algorithm.We effectively proposing a new and sophisticated technique to find morphed images using the features of pHash algorithm

More »»

2019

Gayathri P M, Greeshma K Babu, and G. Deepa, “Optimization of Bug Report Classification using Genetic Algorithm”, International Journal of Recent Technology and Engineering (IJRTE) , vol. 8, no. 1, 2019.[Abstract]


A bug report is an effective way of communicating the bugs among bug reporters and bug recipients. At the same time ,bad bug reports are long, inefficient form of communication for all concerned and do not contain relevant information to resolve the problems .The misclassification in bug report is therefore a serious issue that scarifies the accuracy of bug reports. Here we propose an approach to merging text mining and NLP to identify bugs and non bugs in a bug report. In this system, KNN and Info Gain are used to classify and Genetic algorithm are used to optimize and improve automatic bug prediction performance.

More »»

2018

A. Sunil Kumar, Anand, R., and G. Deepa, “Clustering Online News Comments Using Hadoop On Bigdata”, Journal of Engineering and Applied Sciences, 2018.[Abstract]


Mining in News Blog remarkably a new research area in this modern world of the technological era. Here, we propose a feature word selection of Clustering Online news comments using Hadoop on Bigdata, which realizes structurally superior clustering of online comments. Data is made to run on Hadoop platform so as to convert the unstructured data from the news comments to a structured format for further classification. Here a Naïve Bayesian classifier is included right before applying the K-means clustering algorithm. For clustering, the top most frequent nouns appearing across online comments are selected to construct an overall noun set. Local noun sets are constructed based on the frequently occurring nouns. The global noun set is the intersection of the local and overall noun set. The global noun set is reduced from the corresponding local noun set to construct the distinct noun set.

More »»

2018

K. T. Athira, Gopika, P. G., and G. Deepa, “Collation Between Hierarchical and K- Means Clustering Algorithm”, Journal of Engineering and Applied Sciences, 2018.[Abstract]


Clustering is a technique of keeping the closely related or in other words similar data into groups. Clustering is mainly a process in which a given data set is partitioned into homogenous groups on the basis of certain features like the similar objects are clustered into one group and dissimilar objects are clustered into another group. Therefore, clustering is a type of unsupervised learning and not a supervised learning approach like classification. Different types of clustering techniques are available such as partitioning, density- based, grid-based, hierarchical, model-based and soft-computing methods. In this paper, we propose a comparative study between K-Means and hierarchical clustering methods, claiming that the quality of hierarchical clustering increases as compared to K-Means clustering for the same number of iterations and splitting percentage, as the clustered instances will be more for hierarchical clustering. When performance is considered, K-Means stands over hierarchical clustering. Also we propose that when data is transformed by normalization which is a data preprocessing task results in improved accuracy and quality of Hierarchical clustering as of now.

More »»
Faculty Research Interest: