Predicting Success and Failure in Weight Loss Blogs through Natural Language Use

We explore the emerging phenomenon of blogging about personal goals, and demonstrate how natural language processing tools can be used to uncover psychologically meaningful constructs in blogs. We describe features of a blog community (2638 blogs) devoted to weight loss. We compare several approaches to text analysis in predicting weight loss from natural language use in a subset of the blogs (258 users; over 13,000 entries). First, we use a bag of words approach to distinguish the degree to which individual words can predict success and failure. Next, we compare the results to a deductive word count and categorization tool, Linguistic Inquiry and Word Count. We discuss the theoretical significance of the words and word categories that distinguish between bloggers who succeed and those who fail in their weight loss attempts, along with the implications of automated text analysis in summarizing psychological features of blogs. Blogging about Personal Goals An emerging phenomenon has been the use of blogs for monitoring and documenting the process of self-change (e.g. Harmon, 2003). Thousands of bloggers are chronicling their daily successes and struggles with goals related to weight loss or debt. With each entry, information on weight lost or gained, or on money spent or saved is documented. Examining language use in blogs presents an ecologically valid and unobtrusive way to bypass response rate and selection biases that result when online users are invited to participate in a research survey (see Lyons, Mehl, and Pennebaker 2006). Using text analytic tools, the words that people naturally use to write about the self-change process will be examined. The primary aims of this paper are 1) to find linguistic markers of self-regulatory success and failure in blogs devoted to weight loss, and 2) to compare two text analytic approaches in producing a good predictor model of weight loss success and failure in blogs. ______________________________________________ Copyright © 2008, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Methods Weight Loss Blogs Corpus Publicly available blogs within a community of blogs devoted to weight loss (www.dietdiaries.com) were downloaded, resulting in 2638 blogs. Most bloggers updated their weight with each entry. The community had begun in 1999; the corpus was harvested in November 2007.