Website Evaluation Using Opinion Mining Test phase, srs, design phase and source code final deliverable
Project Domain / Category
Information Retrieval
Abstract /
Introduction
Opinion mining, or sentiment analysis, is a text analysis technique that uses computational linguistics and natural language processing to automatically identify and extract sentiment or opinion from within text (positive, negative, neutral, etc.). Website Evaluation system that rates the website based on the opinion of the user. Stored website(s) will be evaluated based on factors such genuineness of the website, timely delivery of the product after online transaction and support provided by the website. User will comment about the website, based on the comment system will rate and rank the website. The system takes opinion of various users, based on the opinion; system will decide whether the website is genuine or not. The system uses opinion mining methodology in order to achieve desired functionality.
Application developer will use a database of sentiment based keywords along with positivity or negativity weight in database and then based on these sentiment keywords mined in user comment is ranked. The system contains keywords related to fraud, genuineness, timely delivery of the product and service parameters in the database. Based on these factors system will rate the website. With the help of this application user can easily find out which website will deliver the product in time, and also helps to find out website which will provide good support. This application helps to find out whether the website is genuine or not that is useful for those users who do online transactions.
The working of the system is follows:-
- The user logins to the system he can view various websites posted by the admin and can comment about the website.
- User can see the comment of other user.
- System will rate the website based on the comment of various users.
- The role of the admin is to add various website to the system and to add keywords in database.
- So that system will match the comment with the keywords in database and will rate the website based on the sentiment analysis.
Functional
Requirements:
Admin Panel:
- Admin login’s to the system/application by using his/her Admin ID and password.
- Admin can post different topics/item discussion relevant to website.
- Admin will also add various E-Commerce based websites for users.
- Admin will add list of keywords in database relevant to “fraud, genuineness, timely delivery of the product and service” parameters.
User Panel:
- User login’s to the system/application by using his/her user ID and password.
- User can edit his/her profile details along with display picture.
- User will post comments on the topic/website posted by the admin.
- User can also view comment of other users posted on some topic/website.
- Users should be able to comment on stored website/topic only once. The system stores each comments of the users for further processing and find out the sentiments and their weightage and store it in database.
- The stored comments of the users will be analyzed by the system. System will match the user comment with the keywords in database (students will use sentiwordnet dictionary only for this task) and will rate/rank the topic and website.
- User can easily decide whether the stored website(s) provided by the admin/ system are good, bad or worst based on sentiment classification.
User Restrictions:
- Users will not be able to comment more than once on any saved topic/ website by the admin.
- Users will not be able to add any single topic/website (Only admin will add it).
- Users cannot add any keywords in to the system.
Tools:
- SQL 2008
- Visual Studio 2010
- Senti Word Net Dictionary
- Wamp Server Note:
- SentiWordNet Dictionary will be used for sentiment classification tasks only.
- You may use any other platform for coding in which you have command, but make sure all the mentioned FR’s should be completely implemented.