Amelioration of Brand Spam Detection using Live Data

In past few years, user review of any online product or service has been seen as important resource of customer opinion. Existing research has been focused used on extraction, classification and summarization of opinion from reviews in websites, forums and blogs. Now-a-days consumer can obtain information for products and accommodation from online review resources, which can avail them make decision. The social tools provided by the content sharing applications allow online user to interact, to express their opinions and to read opinions from other users. But the spammers provide comments which are written intentionally to mislead users by redirecting them to web sites to increase their rating and to promote products less known on the market. Reading spam comments is a bad experience and a waste of time for most of the online users but can also be harming and cause damage to the reader. Several researchers in this field focused on only fake comments. But, our goal is to detect fake comments which are likely to represent spam considering some indicators like a discontinuous own of text, inadequate and vulgar language or not related to the specific context will helps in giving correct feedback of various customers reviews about given product, Mainly we have observed that previous work is focused on extraction, classification and summarization of opinion and checking of spam and non-spam. But, proposed system aims to evaluate genuine result of filter comments, so that business analyst can make the decision for their organization.