A FRAMEWORK FOR SUMMARIZATION OF ONLINE OPINION USING WEIGHTING SCHEME

Recently there is wide use of social media includes various opinion sites, complaints sites, government sites, question-answering sites, etc. through which customer get services, opinion, information, etc. but because of this there is more and more use of these social media right now so huge amount of data will be created, from this huge data people get confused while taking any decision about particular problem or services. For example, customer wants to purchase a product at that time he/she want the previous customer feedback or opinion about that product. But if there is lots of opinion available for particular product then that customer get confused while taking decision whether purchase that product or not. In this case there is a need of summarization concept means that only show the short and concise manner summary about service or product so that customer or organization easily understand and able to take right decision fast. Our proposed framework creating such summary which contain three main phases or steps. Firstly preprocessing is done in that stop words are removed and stemming is performed. In second phase identify frequent features using two techniques weight constraint and association rule and at the last phase it find semantics and generate the summary so that customer will able to take step without confusion