We present an application of kernel methods to extracting relations from unstructured natural language sources. We introduce kernels defined over shallow parse representations of text, and design efficient algorithms for computing the kernels. We use the devised kernels in conjunction with Support Vector Machine and Voted Perceptron learning algorithms for the task of extracting person-affiliation and organization-location relations from text. We experimentally evaluate the proposed methods and compare them with feature-based learning algorithms, with promising results.

We present an automatic system that discovers competing companies from public information sources. The system extracts data from text, uses transformation-based learning to obtain appropriate data normalization, combines structured and unstructured information sources, uses probabilistic modelling to represent models of linked data, and succeeds in autonomously discovering competitors. We also introduce the iterative graph reconstruction process for inference in relational data, and show that it leads to improvements in performance. We validate system results and deploy it on the web as a powerful analytic tool for individual and institutional investors.