Domain-Specific Aspect Extraction for Product Design

As technology advances, computers become increasingly proficient at interpreting and translating human language into machine-understandable text. With the help of algorithms in natural language processing (NLP), machines can now translate textual data. These algorithms help identify and extract specific text components known as aspects. The aspects represent specific attributes or topics within textual data. For instance, if a product review states,'' This phone has good battery life but poor camera quality,'' and attributes like 'battery life' and 'camera quality' represent aspects in the text. Aspect extraction is a pivotal process involving identifying and isolating key features or topics within text. This research aims to compare and discuss the existing aspect extraction techniques. By effectively extracting the aspects, we will help designers gain the capability to understand and analyze sentiments, thereby enhancing their ability to derive meaningful insights from diverse textual data. Aspect-based sentiment analysis (ABSA) enables the extraction of sentiments towards specific aspects of a product the user provides. For example, when a person writes a review about a restaurant, sentiment analysis can determine whether the review is positive or negative. ABSA can separately determine the review's sentiment towards different aspects of a restaurant, such as, food quality, ambiance, etc. An important input to ABSA are the aspect keywords to find in the reviews. We propose an approach to extract features/aspects from customer-based product reviews. We experiment with different word embedding and clustering techniques to identify the best set of parameters for product design. Our experiments indicate that Global Vectors (GloVe) used on Gaussian Mixture Models (GMMs) yield the most insightful results for the specific problem domain.