Sentiment Analysis in Digital Spaces: An Overview of Reviews

Laura E. M. Ayravainen; Joanne Hinds; Brittany I. Davidson

Sentiment Analysis in Digital Spaces: An Overview of Reviews

Laura E. M. Ayravainen, Joanne Hinds, Brittany I. Davidson

Research output: Working paper / Preprint › Preprint

74 Downloads (Pure)

Abstract

Sentiment analysis (SA) is commonly applied to digital textual data, revealing insight into opinions and feelings. Many systematic reviews have summarized existing work, but often overlook discussions of validity and scientific practices. Here, we present an overview of reviews, synthesizing 38 systematic reviews, containing 2,275 primary studies. We devise a bespoke quality assessment framework designed to assess the rigor and quality of systematic review methodologies and reporting standards. Our findings show diverse applications and methods, limited reporting rigor, and challenges over time. We discuss how future research and practitioners can address these issues and highlight their importance across numerous applications.

Original language	English
Publisher	arXiv
Publication status	Submitted - 30 Oct 2023

Bibliographical note

44 pages, 4 figures, 6 tables, 3 appendices

Keywords

cs.CY
cs.CL
I.2.7

Access to Document

2310.19687v1

Sentiment Analysis Review
Hinds, J. (PI) & Davidson, B. (CoI)
Government Department
1/11/22 → 31/03/25
Project: Central government, health and local authorities

Cite this

@techreport{52547bf7e9a0430d9716b22fabd92934,

title = "Sentiment Analysis in Digital Spaces: An Overview of Reviews",

abstract = "Sentiment analysis (SA) is commonly applied to digital textual data, revealing insight into opinions and feelings. Many systematic reviews have summarized existing work, but often overlook discussions of validity and scientific practices. Here, we present an overview of reviews, synthesizing 38 systematic reviews, containing 2,275 primary studies. We devise a bespoke quality assessment framework designed to assess the rigor and quality of systematic review methodologies and reporting standards. Our findings show diverse applications and methods, limited reporting rigor, and challenges over time. We discuss how future research and practitioners can address these issues and highlight their importance across numerous applications.",

keywords = "cs.CY, cs.CL, I.2.7",

author = "Ayravainen, {Laura E. M.} and Joanne Hinds and Davidson, {Brittany I.}",

note = "44 pages, 4 figures, 6 tables, 3 appendices",

year = "2023",

month = oct,

day = "30",

language = "English",

publisher = "arXiv",

type = "WorkingPaper",

institution = "arXiv",

}

TY - UNPB

T1 - Sentiment Analysis in Digital Spaces

T2 - An Overview of Reviews

AU - Ayravainen, Laura E. M.

AU - Hinds, Joanne

AU - Davidson, Brittany I.

N1 - 44 pages, 4 figures, 6 tables, 3 appendices

PY - 2023/10/30

Y1 - 2023/10/30

N2 - Sentiment analysis (SA) is commonly applied to digital textual data, revealing insight into opinions and feelings. Many systematic reviews have summarized existing work, but often overlook discussions of validity and scientific practices. Here, we present an overview of reviews, synthesizing 38 systematic reviews, containing 2,275 primary studies. We devise a bespoke quality assessment framework designed to assess the rigor and quality of systematic review methodologies and reporting standards. Our findings show diverse applications and methods, limited reporting rigor, and challenges over time. We discuss how future research and practitioners can address these issues and highlight their importance across numerous applications.

AB - Sentiment analysis (SA) is commonly applied to digital textual data, revealing insight into opinions and feelings. Many systematic reviews have summarized existing work, but often overlook discussions of validity and scientific practices. Here, we present an overview of reviews, synthesizing 38 systematic reviews, containing 2,275 primary studies. We devise a bespoke quality assessment framework designed to assess the rigor and quality of systematic review methodologies and reporting standards. Our findings show diverse applications and methods, limited reporting rigor, and challenges over time. We discuss how future research and practitioners can address these issues and highlight their importance across numerous applications.

KW - cs.CY

KW - cs.CL

KW - I.2.7

M3 - Preprint

BT - Sentiment Analysis in Digital Spaces

PB - arXiv

ER -

Sentiment Analysis in Digital Spaces: An Overview of Reviews

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Projects

Sentiment Analysis Review

Cite this