The particular thesis was implemented in the context of my undergraduate studies at the Athens University of Economics and Business, in the Department of Management Science and Technology, during the spring semester 2018–'19. The thesis is a combination of:
- the final report of my internship at isMOOD Data Technology Services company, and
- literature review associated with the project I worked on during this internship.
isMOOD is a Greek B2B company in the field of text analytics,
providing services through an in-house platform and customised reports.
The asset of the company lies in the application of sentiment analysis
on text data coming from various social media sources
(e.g. Facebook, Twitter, Instagram, YouTube, news sites, forums, blogs).
As a Back-end Developer intern at isMOOD I implemented a lexicon-based approach
for performing sentiment analysis on Greek text data.
The technologies I used were Python 2.7 and MongoDB.
Supervisors:
- Academic: Prof. Diomidis Spinellis
- Company: Stauros Triantafyllos (Lead Software Engineer)
An important part of our information-gathering behaviour has always been to find out
what other people think. With the growing availability and popularity of opinion-rich
resources such as online review sites, personal blogs and social media channels, new
opportunities and challenges arise as people now can, and do, actively use information
technologies to seek out and understand the opinions of others. The sudden eruption
of activity in the area of opinion mining and sentiment analysis, which deals with the
computational treatment of opinion, sentiment, and subjectivity in text, has thus occurred
at least in part as a direct response to the surge of interest in new systems that deal directly
with opinions as a first-class object.
This thesis aims to provide a literature review on sentiment analysis, the widespread
available methods for its application to text data, with an emphasis on lexicon-based
approaches for the Greek language. In addition, a particular methodology is proposed and
implemented in the context of an internship, for the application of lexicon-based sentiment
analysis to Greek text data, which facilitates from a well-reputed English sentiment lexicon
by translating it into Greek and transferring sentiment to a Greek lexicon.