Representation of Systematic Literature Review as Heatmaps Based on Sentiment Analysis
Abstract
A systematic literature review (SLR) is a means of identifying, evaluating, and interpreting all available research relevant to a particular research question, topic area, or phenomenon of interest. Currently, the processes for conducting SLR are primarily manual. Therefore, SLR processes require significant time and resources, are limited in scope, and are exposed to errors and bias. the number of publications and publication rates increase each year, which consequently leads to a greater volume of publications that researchers need to examine during a SLR. The current research represents a first attempt at automating the process of SLR through the integration of tools from the field of machine learning. This is through a structured methodology that facilitates the identification of relevant literature, extraction of sentences expressing a position (i.e., a sentiment), and their evaluation in the context of various components included within our domain ontology. Furthermore, this research offers a unique approach to presenting the ontology of the researched field as a matrix of components, which enables a holistic and integrated representation of the sentiments attitude through a heat map. The proposed methodology also contributes to aligning the field of SLR with the principles of open science, facilitating reproducibility by peers, an aspect that is critical for ensuring the objectivity of scientific research.
A systematic literature review (SLR) is a means of identifying, evaluating, and interpreting all available research relevant to a particular research question, topic area, or phenomenon of interest. Currently, the processes for conducting SLR are primarily manual. Therefore, SLR processes require significant time and resources, are limited in scope, and are exposed to errors and bias. the number of publications and publication rates increase each year, which consequently leads to a greater volume of publications that researchers need to examine during a SLR. The current research represents a first attempt at automating the process of SLR through the integration of tools from the field of machine learning. This is through a structured methodology that facilitates the identification of relevant literature, extraction of sentences expressing a position (i.e., a sentiment), and their evaluation in the context of various components included within our domain ontology. Furthermore, this research offers a unique approach to presenting the ontology of the researched field as a matrix of components, which enables a holistic and integrated representation of the sentiments attitude through a heat map. The proposed methodology also contributes to aligning the field of SLR with the principles of open science, facilitating reproducibility by peers, an aspect that is critical for ensuring the objectivity of scientific research.
Last Updated Date : 01/09/2024