Survey on Sentiment analysis on E-commerce

Analysis of sentiment is the process of computationally identifying and categorizing opinions expressed in a pieces of text, especially in order to determine whether the writer’s attitude is positive, negative or neutral towards a particular topic or product. It is the technique due to which we can evaluate the large amount of available data and exact opinions from contain also help customer and organization both to archive their goals.

Now a days the boom of E-commerce site are increasing at very fast rate and they get the huge response from their internet users. After purchasing the product user might share their views and opinion regarding product quality and standards. According to the review, rating is given to the particular product on a star-scaled system.

rating is given to the particular product on a star-scaled system

Sentiment analysis is essential for determine a view point on bases of star, product is evaluated in various categories like most helpful, positive, negative. According to rating the different color of stars are given like green, yellow, red. This topic discuss the need and impact of the sentiment analysis on the E-commerce website. We address the existence of the irrelevant review that effect the customer decision. There are a multiple method’s to archive accuracy of the sentiments analysis. Now we have to move forward with sentiment analysis in electronic text based from which is star with lexicon of positive , negative  and neutral word and phrases like good, nice etc. as a positive and Bed, worst etc. as  a negative.  On bases of this words in comment, rating to the product is give the perfect result and help users to choose most preferable product.


There are a number of methods to evaluate the product and determine that product is good or bed. Naive Bayes is the system which offered the sentiment analyses for language learning on social network platform. Social network includes the twitter here sentiments of words are checked based on word which user write in hashtag and in Facebook its checked on status and user post etc.  This system is evolved to categories an opinion using textual content-degree class and evaluate the content is positive, negative or neutral on the bases of three classifier matrix which are precision, recollect and F-score. Based on sentiment classification Bag-of-words method is used which depends on lexicon extraction from sentiments polarity assignment. This method takes character phrases in a record as function, each word that occur in lexicon has a numerical value like “1” is for positive word and “-1” is for negative .if any word is outside lexicon value then it will consider as “0”.all the above methods are not accurate to giving Exact information about the available online products’. E-commerce website has a great social impact to transforming user’s emotions in text based form and survey on E-commerce sentiment analysis show that the evolution of product is more accurate based on electronic text method.

E-commerce sentiment analyses show that the evolution of product is more accurate based on electronic text method and give’s transparent view to user .It is in feature plan to perform deeper Analysis on the sentiment analysis and recognize the product review on E-commerce sites to promote electronic text based method.

About Author:

Mr.Purav Thakkar is a CEO of one of the fastest growing Offshore Web & Mobile Application Development Company, Innvonix Technologies and well-known blog writer for the mobility solution. He also shares his knowledge about upcoming mobile technologies periodically.