People around the globe are more actively using social media platform such as Twitter, Facebook, and Instagram etc. They share information, opinions, ideas, experiences and other details in the social media. The business communities have become more aware of these developments and they want to use the available information in their favor. One of the ways to understand the people opinions on the product they are using is by collecting tweets related to those products. Then performing the sentiment analysis on the tweets collected on a particular topic.
Understanding customer sentiments is of paramount importance in marketing strategies today. Not only will it give companies an insight as to how customers perceive their products and/or services, but it will also give them an idea on how to improve their offers. This paper attempts to understand the correlation of different variables in customer reviews on a women clothing e-commerce, and to classify each review whether it recommends the reviewed product or not and whether it consists of positive, negative, or neutral sentiment. To achieve these goals, we employed univariate and multivariate analyses on dataset features except for review titles and review texts, and we implemented a bidirectional recurrent neural network (RNN) with long-short term memory unit (LSTM) for recommendation and sentiment classification. Results have shown that a recommendation is a strong indicator of a positive sentiment score, and vice-versa. On the other hand, ratings in product reviews are fuzzy indicators of sentiment scores. We also found out that the bidirectional LSTM was able to reach an F1-score of 0.88 for recommendation classification, and 0.93 for sentiment classification.
Mobile phones have revolutionized the way we purchase products online, making all the information available at our fingertips. As the access to information becomes easier, more and more consumers will seek product information from other consumers apart from the information provided by the seller. Reviews and ratings submitted by consumers are examples of such of type of information and they have already become an integral part of customer's buying-decision process. The review and ratings platform provided by eCommerce players creates transparent system for consumers to take informed decision and feel confident about it. Amazon.com is a treasure trove of product reviews and their review system is accessible across all channels presenting reviews in an easy-to-use format.
Automatically identifying informative reviews is increasingly important given the rapid growth of user generated reviews on sites like Amazon and TripAdvisor. In this paper, we describe and evaluate techniques for identifying and recommending helpful product reviews using a combination of review features, including topical and sentiment information, mined from a review corpus.
While sentiment detection, identification, and classification are popular research areas, researchers frequently work in only one domain at a time. Typical domains include movie reviews (Pang et al. 2002) and product reviews (Dave et al. 2003). Performing sentiment detection upon keywords chosen at run time is more difficult. The techniques applied to determine the sentiment of keywords in movie and product reviews are less effective when used on blogs due to a variety of reasons. Unlike reviews blogs tend to talk about many different subjects at a time making many NLP and machine learning approaches more difficult. Finally, many of the techniques used in the different review domains incorporate domain specific knowledge. The 2006 NIST TREC Blog track (Ounis et al. 2006) on "opinion retrieval" from blog posts, presents an opportunity to tackle this problem. The task was defined as follows: build a system that will take a query string describing a topic, e.g., "March of the Penguins", and return a ranked list of blog posts that express an opinion, positive or negative, about the topic. NIST provided researchers with a data set of over three million blogs, and judged entries upon retrieval results for a set of fifty test queries.