Identifying User Needs from Social Media

With the rise of social media, writings by ordinary people are becoming increasingly available for linguistic analysis. Such analyses offer great opportunities to identify individual users’ needs from user-generated content, so that better tailored products or services can be recommended. Literature suggests that several types of human needs are universal and directly influence consumer purchase behavior. In this paper, we investigate the use of social media to identify such fundamental needs for individuals. We developed psychometric measures of universal needs through a crowd-sourced study. We also built several models to predict people’s needs based on their writings. We conducted a detailed analysis of the models and showed that our models can effectively identify users’ needs based on their social media data. Our results also confirm that some inferred needs correlate well with the actual product purchases and suggest a great potential for our models to significantly increase effectiveness of product recommendations. Author