Opinion-Driven App Recommender System (ODARS)
Name
Shalva Avanashvili
Abstract
Currently, there are no decent ways for the users to quickly determine what people are thinking about the specific application and its features, or which application is better than the other. Users mainly rely on the ratings, some articles or reviews before downloading the application. Unfortunately, it is really difficult for the human to go through all reviews in order to get an impression on an application, to see how positively or negatively people have been thinking about the specific features.
Recently, a SAFE approach was proposed for app feature extraction from user reviews and app description. The approach has shown its superiority over the previous techniques. However, there is no tool that that uses the recent SAFE approach to analyze user sentiments mentioned in user reviews at app feature-level.
The intention of this study is to develop a tool to analyze user sentiments mentioned on app features in user reviews. The tool enables to perform analysis at single app or multiple apps levels. The app features are extracted from the user reviews and app description together. A survey is conducted to evaluate the developed tool based on its ease of use, usefulness and future use.
Recently, a SAFE approach was proposed for app feature extraction from user reviews and app description. The approach has shown its superiority over the previous techniques. However, there is no tool that that uses the recent SAFE approach to analyze user sentiments mentioned in user reviews at app feature-level.
The intention of this study is to develop a tool to analyze user sentiments mentioned on app features in user reviews. The tool enables to perform analysis at single app or multiple apps levels. The app features are extracted from the user reviews and app description together. A survey is conducted to evaluate the developed tool based on its ease of use, usefulness and future use.
Graduation Thesis language
English
Graduation Thesis type
Master - Software Engineering
Supervisor(s)
Faiz Ali Shah, Dietmar Pfahl
Defence year
2018