Sentiment analysis using machine learning
Leuhu, Tiia (2015)
2015
Signaalinkäsittelyn ja tietoliikennetekniikan koulutusohjelma
Tieto- ja sähkötekniikan tiedekunta - Faculty of Computing and Electrical Engineering
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2015-06-03
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201505201399
https://urn.fi/URN:NBN:fi:tty-201505201399
Tiivistelmä
In recent years, social media and TV-production has formed a strong link between each other. The most popular social media platform in TV-industry is Twitter, where over a million tweets are shared in one day. Tweet content is feedback straight from the viewers, and might include more valuable information than individual surveys. Going through millions of tweets is hard or impossible manually. This thesis studies, how to teach a machine by supervised manner to analyze tweets. Machine analyzes sentiments based on the features that tweets include.
The main goal of this thesis is to clarify how the content can be received, prepared, extracted and classified. The study indicates that sentiments can be caught from Twitter data using mathematical patterns.
The thesis is divided into 5 chapters. Chapter 1 is the introduction for the sentiment analyzing with machine learning capabilities. Chapter 2 is the literature study part, where elements and techniques are explored. Chapter 3 is the implementation part, where selected classification methods and techniques for text data are specified. Chapter 4 covers results and chapter 5 finishes the work with conclusions.
The main goal of this thesis is to clarify how the content can be received, prepared, extracted and classified. The study indicates that sentiments can be caught from Twitter data using mathematical patterns.
The thesis is divided into 5 chapters. Chapter 1 is the introduction for the sentiment analyzing with machine learning capabilities. Chapter 2 is the literature study part, where elements and techniques are explored. Chapter 3 is the implementation part, where selected classification methods and techniques for text data are specified. Chapter 4 covers results and chapter 5 finishes the work with conclusions.