Quantifying product favorability and extracting notable product features using large scale social media data

Some of the challenges that designers face in getting broad external input from customers during and after product launch include geographic limitations and the need for physical interaction with the design artifact(s). Having to conduct such user-based studies would require huge amounts of time and financial resources. In the past decade, social media has emerged as an increasingly important medium of communication and information sharing. Being able to mine and harness product-relevant knowledge within such a massive, readily accessible collection of data would give designers an alternative way to learn customers' preferences in a timely and cost-effective manner. In this paper, we propose a data mining driven methodology that identifies product features and associated customer opinions favorably received in the market space which can then be integrated into the design of next generation products. Two unique product domains (smartphones and automobiles) are investigated to validate the proposed methodology and establish social media data as a viable source of large scale, heterogeneous data relevant to next generation product design and development. We demonstrate in our case studies that incorporating suggested features into next generation products can result in favorable sentiment from social media users.