Objective:
Recommendation systems provide an approach to facilitate the user’s desire. It helps recommend things from various domains. Researchers express their ideas and experience in an academic article for the research community. However, they have ample options when they aspire to publish. At times, they end up with incorrect submissions resulting in a waste of time and effort for the editor and himself. Journal selection has been a very tedious task for novice authors.
Molecular Connection’s Solution:
Molecular Connection started by evaluating many freely available solutions and found that most solutions needed to provide the correct recommendation. Most of the solutions were based on the keyword-based approach rather than understanding the context of the paper and suggesting the most appropriate journal.
After conducting extensive user research, Molecular Connections developed a Deep learning enabled module, which considers
The scope of the paper
The underlying published papers
The past rejections
The changing publication trends
The system ranks any new manuscript and provides a suitability score against a set of journals.
Equipped with a self-learning module, the system evolves based on user activity and aligns with changes.
Integration:
The JRS system is built using the content available from the publisher. Publishers have multiple options for integrating the system. SAAS, On-premises are a few of the many options available,
Key Business Benefits:
User Engagement for authors
Journal Cascading for Editors
Many researchers are faced with an overwhelming decision about where to publish their research. While there are many great journals out there, which ones are the best? That’s a hard question to answer, and it often seems as though there are just too many choices! On the other hand, editors and editorial staff are faced with a similar dilemmas of balancing out the manuscript review process by considering the scope of the journals in the portfolio, the time taken for a manuscript for approval by reviewers, deciding on implementing a transfer desk that ensures that no incoming content with value is misdirected or passed up. Our Journal Recommendation System (JRS) solves both ends of the problems posed for both authors and editorial staff by allowing them to make these decisions based on heuristically backed, efficient suggestions. It is powered by state-of-the-art deep learning algorithms and decision trees.
