The foremost goal of this research is to analyze the reviews of E-Commerce products using Social Media Big Data using proposed Rule Based Classification Technique. This proposed work is about evaluating customer's Tweets about Amazon products. Customer's Tweets may vary according to the person's taste and style. And Amazon reviews presented in the website is very huge this is making it impossible for customers to discover about product reaction. This issue is addressed in the proposed approach, which involves undertaking opinion mining across numerous social networks. Its goal is to gather reviews from Twitter in order to scrutinize Amazon product sentiment. And it delivers useful information to online shopping customers about product worth, as well as valuable decision-making thoughts to businesses about the customer's favorite for and buys products. This work take into account of various kinds of opinion words found in tweets, such as capitalized words, repeated letter sequences, negation words, modifier items, emoji, intensifiers, slang terms, conjunction words and exclamatory words. This proposed work integrated the five kinds of sentiment or lexicon dictionary. This kind of amalgamation can reduce the lacking of opinion words, and also can avoid the missing polarity value.