Integrating Probabilistic Extraction Models and Data Mining to Discover Relations and Patterns in Text

In order for relation extraction systems to obtain human-level performance, they must be able to incorporate relational patterns inherent in the data (for example, that one\u27s sister is likely one\u27s mother\u27s daughter, or that children are likely to attend the same college as their parents). Hand-coding such knowledge can be time-consuming and inadequate. Additionally, there may exist many interesting, unknown relational patterns that both improve extraction performance and provide insight into text. We describe a probabilistic extraction model that provides mutual benefits to both ``top-down\u27\u27 relational pattern discovery and ``bottom-up\u27\u27 relation extraction