Evaluating Cross Domain Sentiment Analysis using Supervised Machine Learning Techniques

Azwa Abdul Aziz; Andrew Starkey; Marcus Campbell Bannerman

doi:10.1109/IntelliSys.2017.8324369

Evaluating Cross Domain Sentiment Analysis using Supervised Machine Learning Techniques

Azwa Abdul Aziz, Andrew Starkey, Marcus Campbell Bannerman

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

14 Citations (Scopus)

Abstract

Sentiment Analysis is the process of computationally identifying and categorizing opinion expressed in a piece of text, especially in order to determine whether the writer's attitude towards a particular topic is negative, positive or neutral. Many researchers have proposed novel methods for sentiment classification especially using supervised machine learning (ML) techniques. However, there is still limited research with successful results in Cross-Domain Sentiment Analysis. Therefore, previous experiments were replicated by using different ML techniques with several enhancements in order to better understand the sentiment classification process and to compare results with cross-domain analysis. Limitations of the proposed approach are discussed and a new automated model is suggested for future work.

Original language	English
Title of host publication	Intelligent Systems Conference 2017
Place of Publication	London
Publisher	IEEE Explore
Number of pages	8
ISBN (Electronic)	978-1-5090-6435-9
ISBN (Print)	978-1-5090-6436-6
DOIs	https://doi.org/10.1109/IntelliSys.2017.8324369
Publication status	Published - 7 Sept 2017
Event	SAI Intelligent Systems Conference 2017 (IntelliSys 2017) - America Square Conference Center, London, United Kingdom Duration: 7 Sept 2017 → 8 Sept 2017

Conference

Conference	SAI Intelligent Systems Conference 2017 (IntelliSys 2017)
Country/Territory	United Kingdom
City	London
Period	7/09/17 → 8/09/17

Keywords

cross-domain sentiment analysis
machine learning
supervised techniques
Lexicon-Based Approach

Access to Document

10.1109/IntelliSys.2017.8324369

Cite this

@inproceedings{a3eb6696f8b2452db2e3d31a7f277941,

title = "Evaluating Cross Domain Sentiment Analysis using Supervised Machine Learning Techniques",

abstract = "Sentiment Analysis is the process of computationally identifying and categorizing opinion expressed in a piece of text, especially in order to determine whether the writer's attitude towards a particular topic is negative, positive or neutral. Many researchers have proposed novel methods for sentiment classification especially using supervised machine learning (ML) techniques. However, there is still limited research with successful results in Cross-Domain Sentiment Analysis. Therefore, previous experiments were replicated by using different ML techniques with several enhancements in order to better understand the sentiment classification process and to compare results with cross-domain analysis. Limitations of the proposed approach are discussed and a new automated model is suggested for future work.",

keywords = "cross-domain sentiment analysis, machine learning, supervised techniques, Lexicon-Based Approach",

author = "{Abdul Aziz}, Azwa and Andrew Starkey and {Campbell Bannerman}, Marcus",

year = "2017",

month = sep,

day = "7",

doi = "10.1109/IntelliSys.2017.8324369",

language = "English",

isbn = "978-1-5090-6436-6",

booktitle = "Intelligent Systems Conference 2017",

publisher = "IEEE Explore",

note = "SAI Intelligent Systems Conference 2017 (IntelliSys 2017) ; Conference date: 07-09-2017 Through 08-09-2017",

}

TY - GEN

T1 - Evaluating Cross Domain Sentiment Analysis using Supervised Machine Learning Techniques

AU - Abdul Aziz, Azwa

AU - Starkey, Andrew

AU - Campbell Bannerman, Marcus

PY - 2017/9/7

Y1 - 2017/9/7

N2 - Sentiment Analysis is the process of computationally identifying and categorizing opinion expressed in a piece of text, especially in order to determine whether the writer's attitude towards a particular topic is negative, positive or neutral. Many researchers have proposed novel methods for sentiment classification especially using supervised machine learning (ML) techniques. However, there is still limited research with successful results in Cross-Domain Sentiment Analysis. Therefore, previous experiments were replicated by using different ML techniques with several enhancements in order to better understand the sentiment classification process and to compare results with cross-domain analysis. Limitations of the proposed approach are discussed and a new automated model is suggested for future work.

AB - Sentiment Analysis is the process of computationally identifying and categorizing opinion expressed in a piece of text, especially in order to determine whether the writer's attitude towards a particular topic is negative, positive or neutral. Many researchers have proposed novel methods for sentiment classification especially using supervised machine learning (ML) techniques. However, there is still limited research with successful results in Cross-Domain Sentiment Analysis. Therefore, previous experiments were replicated by using different ML techniques with several enhancements in order to better understand the sentiment classification process and to compare results with cross-domain analysis. Limitations of the proposed approach are discussed and a new automated model is suggested for future work.

KW - cross-domain sentiment analysis

KW - machine learning

KW - supervised techniques

KW - Lexicon-Based Approach

U2 - 10.1109/IntelliSys.2017.8324369

DO - 10.1109/IntelliSys.2017.8324369

M3 - Published conference contribution

SN - 978-1-5090-6436-6

BT - Intelligent Systems Conference 2017

PB - IEEE Explore

CY - London

T2 - SAI Intelligent Systems Conference 2017 (IntelliSys 2017)

Y2 - 7 September 2017 through 8 September 2017

ER -

Evaluating Cross Domain Sentiment Analysis using Supervised Machine Learning Techniques

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this