Gerard Deepak

Gerard Deepak

Auto-generated profile, based on publicly available information

Gerard Deepak

Department of Computer Science and Engineering, Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, India

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2018 in Procedia Computer Science
Verified
Journal Article
Open access
Zameer GulzarZameer Gulzar
Zameer Gulzar
AAAnny Leema A
Anny Leema A
GDGerard Deepak
Gerard Deepak

6 September 2018 in Computers and Electrical Engineering
Verified
Journal Article
Closed access
GDGerard Deepak
Gerard Deepak
JPJ Sheeba Priyadarshini
J Sheeba Priyadarshini

1 January 2019 in International Journal of Computer Aided Engineering and Technology
Verified
Journal Article
Closed access
GDGerard Deepak
Gerard Deepak
DKDheera Kasaraneni
Dheera Kasaraneni
With the paradigm shift in business strategy in terms of online marketing and e-commerce and to comprehend the World Wide Web transforming into an intelligent semantic web, there arises a perpetual need for semantically driven e-commerce systems which gives preference to the users. In this paper, OntoCommerce which is an e-commerce system that incorporates semantic algorithms for product recommendation has been proposed. The proposed strategy uses the enriched normalised pointwise mutual information measure for semantic similarity computation. OntoCommerce assimilates ontologies and recommends products based on the user query, recorded user navigation as well as the user profile analysis thereby encompassing personalisation. In order to make the recommendations more relevant, OntoCommerce uses parametric fuzzification to increase the number of relevant recommendable products. OntoCommerce yields an average accuracy of 88.68 % with a low false discovery rate of 0.13 which makes it a best-in-class semantically driven product recommendation system for online e-commerce.

2 January 2020 in Journal of Discrete Mathematical Sciences and Cryptography
Verified
Journal Article
Closed access
GDGerard Deepak
Gerard Deepak
VTVarun Teja
Varun Teja
ASA. Santhanavijayan
A. Santhanavijayan
In this paper, contemporary Natural Language Processing techniques have been leveraged to demonstrate the capability of data-driven HR towards significant improvement in the quality and speed of the whole recruiting process. Firstly, by using NLP, a resume parser has been implemented to analyze the most crucial recruitment parameters. Thereafter, ability to display a pie chart for a candidate has been employed in the algorithmic structure of the parser to prepare a powerful tool for the resume matching based on job criteria. To determine the efficacy and accuracy of the proposed resume ranker, an enhanced rival modern optimizer, i.e., firefly ranking algorithm is applied to accelerate the speed of ranking algorithm. An overall accuracy of 94.19% has been achieved by the proposed approach. The results indicate that the resume parser has been incorporated with robust techniques and hence concedes to the accuracy of the results.

28 May 2021 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
GDGerard Deepak
Gerard Deepak
SRS. Rooban
S. Rooban
ASA. Santhanavijayan
A. Santhanavijayan

1 December 2019
Verified
Conference Paper
Closed access
GDGerard Deepak
Gerard Deepak
NKNaresh Kumar
Naresh Kumar
GBG Vsn Sai Yashaswea Bharadwaj
G Vsn Sai Yashaswea Bharadwaj
ASA Santhanavijayan
A Santhanavijayan
In modern times, e-learning has become the most convenient environment to attain knowledge about and across various domains. While there are plenty of e-learning resources available across the ever-expanding web, imbibing knowledge is not the only factor that contributes to the edification of an individual. The assessment also must be given a quintessential role because both learning and assessment are two sides of a coin. In this paper, OntoQuest, which is a scheme for generation of multiple-choice questions is done based on the domain or subject of choice from the user. An equivalence set based summarization scheme has been proposed. Synonymization and Anonymization are used for the generation of best key and distractors respectively. Domain and Granular Ontologies are crawled for relevant sub-topics and auxiliary topics. WordNet integrates dynamic knowledge for improving the overall accuracy. Jaccard similarity is encompassed for the generation of the correct key. OntoQuest is a reliable framework that yields an overall accuracy of 95.46% and 95.05% for key and distractor generation respectively, which is higher and consistent compared to the existing models available.