An Assisted Literature Review using Machine Learning Models to Identify and Build a Literature Corpus

Journal Title: International Journal of Engineering and Science Invention - Year 2017, Vol 6, Issue 7

Abstract

With the evolving and interdisciplinary nature of research, there is a need to facilitate and assist researchers in the manual process of building a literature review. This paper proposes an assisted literature review prototype based on machine learning models (MLM) to discover, rank and recommend the relevant papers. Using text and data mining models, MLM and a classification model, all of which learn from researchers’ annotated data and semantic enriched metadata, assisted literature review helps researchers identify, rank and select papers. This prototype evaluates papers and bibliographic attributes in order to determine their relevancy and aggregates all relevant contents into an assisted literature review object. This paper presents the MLM and related algorithms that: 1. Identify the relevant papers harvested from the web and other sources for building the Literature Corpus; 2. Obtain the Literature Corpus radius by calculating the distance of each paper to the center of the Literature Corpus defined for a specific topic, concept or area of research. The performance, in terms of accuracy, was evaluated and compared to other approaches using a number of assisted literature review prototype simulations with a manual literature review metadata as input parameters.

Authors and Affiliations

Ronald Brisebois, Alain Abran, Apollinaire Nadembega, Philippe N‘techobo

Keywords

Related Articles

Development of NaCl Salt as a Reusable Gamma Radiation Dosimeter

NaCl salt is a well-known material that can be used just one time for retrospective dosimetryin radiation accidents. After first time of use, the sensitivity of NaCl drops dramatically and its TL glow curve changes signi...

Architecture: A Novel Agenda

Today Architecture cannot ignore the hard realities of problems of a burgeoning population; its meager resources or its limited ability to pay for a shelter but in the evolution of every society, beyond the physiological...

Design Thinking Plausible Ancient Indian Alchemical Topical Drug Mechanism for Healing Wounds – A Mechanistic Approach

Vanishing ancient technologies have in them a vast treasure of mystique methods or techniques that have to be researched to verify their compatibility with modern technologies. Of them, Alchemy, especially Indian alchemy...

A Survey on Security of Iot devices

IoT devices industry is rapidly growing with an ever-increasing list of manufacturers offering a countless number of smart devices targeted to enhance end-user’s standard of living. Security is an afterthought in these d...

Review: Industrial Internet of Things (IIoT) in Agriculture

Industrial Internet of Things (IIoT) is the utilization of Internet of Things (IoT) innovations in manufacturing. The thought behind IIoT is that the technologies ought to interface things to infrastructure. By this, it...

Download PDF file
  • EP ID EP403274
  • DOI -
  • Views 48
  • Downloads 0

How To Cite

Ronald Brisebois, Alain Abran, Apollinaire Nadembega, Philippe N‘techobo (2017). An Assisted Literature Review using Machine Learning Models to Identify and Build a Literature Corpus. International Journal of Engineering and Science Invention, 6(7), 72-84. https://europub.co.uk/articles/-A-403274