Contact at or 8097636691
Responsive Ads Here

Sunday, 17 June 2018



Due to the increase in demand for e-commerce with people preferring online purchasing of goods and products, there is a vast amount of information being shared. The e-commerce websites are loaded with large volume of data. Also, social media helps a great deal in sharing of this information. This has greatly influenced consumer habits all over the world. Due to the vivid reviews provided by the customers, there is a feedback environment being developed for helping customers buy the right product and guiding companies to enhance the features of product suiting consumer’s demand. The only disadvantage of availability of this huge volume of data is its diversity and its structural non-uniformness. The customer finds it difficult to precisely find the review for a particular feature of a product that s/he intends to buy. Also, there is a mixture of positive and negative reviews thereby making it difficult for customer to find a cogent response. Also these reviews suffer from spammed reviews from unauthenticated users. So to avoid this confusion and make this review system more transparent and user friendly we propose a technique to extract feature based opinion from a diverse pool of reviews and processing it further to segregate it with respect to the aspects of the product and further classifying it into positive and negative reviews using machine learning based approach.

KEYWORDS: aspect; sentiment analysis; feature extraction; machine learning


 In the recent years E-Commerce has exploded everywhere in the world, and the majority of the population is preferring to buy products through these websites. Consequently large amount of data in the form of reviews is produced which helps prospective buyers to choose the right product. Furthermore these reviews contain opinionated contents which can be useful for the company to identify the areas which need to be enhanced. However it is impractical for the user to read each and every review about the product. Moreover, reading only few reviews may present a biased idea about the product. It is quite possible that some of the reviews lack credible sources, which the users have no means to differentiate. Besides the reviews and ratings provided do little to assess the specific features of the product. Due to all the above constraints, the user is unable to make a fully informed decision about the product.

Opinion mining also known as sentiment analysis can be used to extract customer reviews from different sources on the internet. This technique implements various algorithms to analyze the corpus of data and make sense out of it. This technique helps to identify the orientation of a sentence thereby recognising the element of positivity or negativity in it. Automated opinion mining can be implemented through a machine learning based approach. Opinion mining uses natural language processing to extract the subjective information from the data (in this case it’s customer reviews).

the general public towards a particular thing, person or an event. There are three general levels for opinion mining tasks: document level, sentence level and phrase level in Liu[1]. Document level tasks mainly help in segregating the overall document into either subjective document or objective document. Further it can be distinguished into positive, negative or neutral. It can also help separate the spam from the non spam. The sentence level opinion mining is performed on the sentences which can help group certain sentences to summarise the opinion and also it can help identify comparative sentences to rank them accordingly. Phrase level deals with the aspects and is known as aspect based opinion mining. This helps to identify the reviewers sentiment about specific aspects of the product. This level does the finer-grained analysis of the opinions.

No comments:

Post a Comment