Ανάλυση συναισθημάτων σύντομων μηνυμάτων σε εφαρμογή Android με χρήση τεχνολογιών Google Cloud Messaging και Data Mining (Μέρος Α)

Παρασκάκης, Ιωάννης Χρ.

Sentiment analysis of instant messages on android application using Google Cloud Messaging technologies and Data Mining (Part A)

Master Thesis

Author

Παρασκάκης, Ιωάννης Χρ.

Date

2016-09

Abstract

Millions of people around the world use, on a daily basis, microblogging sites and instant messaging applications in order to communicate, exchange ideas and opinions. This phenomenon has attracted the interest of the research community in the realm of sentiment analysis of these interactions. The evolution of the handheld devices and their increasing capabilities (in terms of sensors and hardware) produces a vast amount of instant message meta-data which are available and can be used to better understand the context (environmental and sentimental) around the use of the aforementioned applications and services. Sentiment analysis, also known as opinion mining is the area of science that studies the opinions, feelings, evaluations, assessments, attitudes and feelings of the people to entities such as products, services, organizations, people, events, social issues and their characteristics. The analysis of emotions as an area of research is large and relatively new and unexplored. The first mention perhaps the term found in the not too distant, for research data 2003 (Jeonghee Yi, 2003). The task of inference from user interactions in online social networks (OSN) and the analysis of natural language (NLP) becomes even more difficult if we consider the range of available languages and dialects and the lack of organized and comprehensive data sets in combination with the plethora of emotions that can be hidden in a small text of 140 characters (Twitter). The aim of this thesis is to create an integrated platform for exchanging short messages using Cloud technologies (Google Cloud Messaging). Furthermore the aforementioned platform will incorporate an evaluation and categorization system of messages in emotion classes using modeling and classification techniques build with the use of RapidMiner tool. It is thus an integrated transmission and evaluation of small text message that includes a user application for mobile Android devices, an application server that manages the storage and transmission of messages through the Google Cloud Messaging and a message evaluation model.

Postgraduate Studies Programme

Προηγμένα Συστήματα Πληροφορικής

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Πληροφορικής

Number of pages

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/9745

Collections

Τμήμα Πληροφορικής

Show full item record

Except where otherwise noted, this item's license is described as
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές