Section: New Results

Expert Opinion Extraction from a Biomedical Database

Participants : Ahmed Samet, Thomas Guyet, Benjamin Négrevergne, Tien-Tuan Dao, Tuan Nha Hoang, Marie-Christine Ho Ba Tho.

This work tackles the problem of extracting frequent opinions from uncertain databases. This problem is encountered in real-world applications, such as the opinions of medical experts to evaluate the reliability level of biomedical data. We introduce the foundation of an opinion mining approach with the definition of pattern and support measure. The support measure is derived from the commitment definition. In [15], we proposed a new algorithm called OpMiner that extracts the set of frequent opinions modeled as a mass functions. We applied it on a real-world biomedical opinion database. Performance analysis showed that our proposal generated better patterns compared to literature-based methods.