Case Studies

Our client is a marketing agency helping businesses to market their products on major marketplaces like Amazon. Since online shopping grew in popularity exponentially in the last couple of years, the competition within the online marketplaces has grown to the point where just putting the product online is not enough to achieve high sales. Product marketing now can give businesses a crucial advantage over their competitors, and a consumer-driven marketing campaign can significantly boost sales.

Our client has approached us to create a system that would analyze Amazon reviews and show the customers' feedback about the product in a form of product aspects, i.e. customer opinions grouped by topic:

• what aspects have they found to be the most satisfying about the product
• what disappointed them the most about the product
• aspects they have found surprising, interesting, etc.

The customer feedback pulled from the reviews in a concise form can help create convincing marketing materials, like product descriptions, articles, etc., as well as study competitors' weak points and conduct customer-based market research.

The system we have developed uses Amazon API to extract Amazon reviews for any given product. After the reviews are extracted, the system splits reviews into sentences, groups them by topic, and chooses a sentence that best describes the topic. Results are presented in a form of aspects - main themes extracted from reviews. Each aspect can be expanded to see sentences from which the aspect was extracted.

The system processes hundreds of reviews and thousands of sentences for each product to extract the most relevant aspects, ignoring sentences that don’t contain relevant information:

• personal opinions - ‘I love it’, ‘Worst phone I’ve ever used’
• exclamations - ‘Wow’, ‘Great’
• comments about customer service or marketplace: ‘Amazon lost my package’, ‘customer support was really helpful’

Sentence processing is done using multiple NLP and machine learning models using GPU for training. The system can correctly process natural language, including slang, abbreviations, and contractions.

We have leveraged the power of ChatGPT to extract main characteristics of a product. The reviews are separated into multiple groups and are fed to ChatGPT with a prompt asking it to describe top 10 strongest and weakest aspects of the product, which are then extracted into our app. This approach to aspect extraction has provided superior results in lesser time.

Our NLP system is capable of extracting product aspects from Amazon reviews, sorting them by relevance, and showing how users describe the product, helping our client to create convincing, user-centric marketing materials and increase product sales. The system can be used as a standalone app or integrated into an existing application for the purposes of customer sentiment analysis, marketing research, increasing customer satisfaction through providing review summaries, etc.