An efficient framework for scientific article recommendation system

Title

An efficient framework for scientific article recommendation system

Creator

Nair, Akhil M.

Contributor

George, Jossy P.

Description

Excess data makes it challenging to extract information that is relevant to a domain of study or research. Existing state-of-the-art systems focus majorly on the selection of highly connected, prestigious and cited articles, regardless of the relevance of papers. To improve quality of findings, recommender systems which are a subclass of information filtration systems are used. They filter out relevant information over prestigious data from an existing repository of information. There are various sub-domains under recommender systems. This study focuses on citation recommendation. Citations are an integral part of any scientific paper, academic dissertation or projects. Finding appropriate citations for any work is a scholar's most time-consuming task. Thus, a well-defined citation recommendation system provides fulfillment and completeness for citing the giants works. The thesis aims to study existing frameworks for citation recommendation systems and identify the best dataset to work on graph- based recommender systems. A framework that recommends the most similar and relevant article to the user rather than prestigious authors or papers is here by proposed. The study explores various machine learning and deep learning techniques and methods which can be used effectively in recommending loosely connected yet highly relevant articles.

Source

Author's Submission

Date

2022-01-01

Publisher

Christ(Deemed to be University)

Subject

Computer Science

Rights

Open Access

Relation

No Thesis

Format

PDF

Language

English

Type

PhD

Identifier

http://hdl.handle.net/10603/426631