What Sentiments Can Be Found in Medical Forums?

Proceedings of Recent Advances in Natural Language Processing, pages 633–639, Hissar, Bulgaria, 7-13 September 2013. What Sentiments Can Be Found in Medical Forums? Marina Sokolova CHEOR Research Institute University of Ottawa sokolova@uottawa.ca Victoria Bobicev Technical University of Moldova vika@rol.md Abstract In this work we present sentiment analysis of messag- es posted on a medical forum. We categorize posts, written in English, into five categories: encourage- ment, gratitude, confusion, facts, and facts + senti- ments. Our study applies a manual sentiment annota- tion, affective lexicons in its sentiment analysis and machine learning classification of sentiments in these texts. We report empirical results obtained from anal- ysis of 752 posts dedicated to infertility treatments. Our best results improve multi-class sentiment classi- fication of online messages (F-score = 0.518, AUC= 0.685). 1 Introduction User-friendly Web 2.0 technologies encourage the general public actively participate in the crea- tion of the Web content. Blogs, social networks, message boards reach out to a global community of the Web users. The online texts discuss per- sonal experience and convey sentiments and emotions of the authors. These emotion-rich posts are known to be important in setting inter- action patterns among members of online com- munities as emotion-rich text has a strong influ- ence on a public mood (Allan, 2005). Subjective information posted by a user may affect subjec- tivity in posts written by other users (Zafarani et al 2010). Studies of online sentiments and opinions can help in understanding of sentiments and opinions of the public at large. Such understanding is es- pecially important for the development of public policies whose success greatly depends on public attitudes. Among major policy issues (e.g., edu- cation, internal and foreign affairs), health care policies are those that directly affect everybody and cause many online discussions. A 2011 sur- vey of the US population estimated that 59% of all adults have looked online for information about health topics such as a specific disease or treatment (Fox 2011). Reproductive technolo- gies belong to a group of hotly debated health care issues in the modern societies (Zillen 2011). The systematic review of 19 studies from 1999- 2009 listed several reasons for the use of medical forums: a) information searching - to learn about psychological, physical and social aspects of available treatments, evaluations of alternative treatments; b) in seeking emotional support - anonymous communication, immediate and con- stant community access, easy contact to peers. We analyzed sentiments expressed by partici- pants of In Vitro Fertilization (IVF) medical fo- rum. 1 This forum brings together women who use IVF treatments with the hope to conceive. For the empirical analysis, we selected 752 posts that covered 74 topics related to IVF (e.g., Over 40 and pregnant or trying to be, Odds of getting pregnant naturally on a cancelled IVf cycle, Go- ing for a second opinion). Starting with several possible sentiments, we finally categorized text into encouragement, gratitude, confusion, facts + encouragement, and facts. Texts in which the annotators disagreed on a class label were la- beled …

Content truncated to 500 words