Architecture of the Sentiment Web Mining

In this paper we had described the system architecture which is still in the development process whose purpose is to collect the sentiment of web users regarding various topic such as social issues like immigration, retail products, and financial instruments (FI).

The first step is acquiring of knowledge or knowledge acquisition. Various sources on the web helps a Sentiment Web Mining (SWM) system to acquire knowledge. Blogs, social networks, email, or online news proves the main place for finding such knowledge. A SWM system has personalization and customization capabilities. Customization occurs when the SWM user can change his/her preferences as it helps to select specific sites that can be used for data mining and evaluation. Based on the user profile, when the system decides which sites to be used for data mining, personalization occurs. The second step is storage of knowledge, which involves database creation. Indexing and tagging of appropriate web sites takes place. The hardest part of this step is Tax anatomy. A SWM system will use a series of off–the-shelf knowledge analysis/data mining tools including SWM knowledge analysis/data mining engine which is based on web services technology. The last step is widely spreading of knowledge to the users. The presentation component of a SWM system is separated from other components, namely, the process component, data access component, and business rule component so that they can be maintained easily.