A Framework for Opinion Mining System with Design Pattern

Abstract

Due to the sheer volume of opinion rich web resources such as discussion forum, review sites, blogs, and news corpora available in digital form, much of the current research is focusing on the area of sentiment analysis. People are intended to develop a system that can identify and classify opinion or sentiment as represented in an electronic text. An accurate method for predicting sentiments could enable us, to extract opinions from the internet and predict on-line customer's preferences, which could prove valuable for economic or marketing research. In this paper we present a framework for opinion mining in Traditional Chinese-called FOM (Framework of Opinion Mining) to collect unstructured articles in the popular web site and analyse the opinion and sentiment in the semi-automatic way. The framework is developed by objected oriented design patterns, such as to support the flexibility and maintainability. With the FOM framework, new analysis algorithm can be easily replaced and integrated in a new application. A flood predication application based on facebook text in Taiwan will be demonstrated in this paper.