Discourse Analyst
At H5 I built conceptual and linguistic models for information retrieval systems. Our systems were designed to automate processes related to document review (for litigation) and document retention (for legal compliance). Building models required specialized expertise, such as in linguistic analysis, natural language processing, ontology development, and regular expressions.On client projects, I worked with attorneys and other linguists to define clear, nuanced subject-matter categories… Show more At H5 I built conceptual and linguistic models for information retrieval systems. Our systems were designed to automate processes related to document review (for litigation) and document retention (for legal compliance). Building models required specialized expertise, such as in linguistic analysis, natural language processing, ontology development, and regular expressions.On client projects, I worked with attorneys and other linguists to define clear, nuanced subject-matter categories based on legal requirements, and to regularly review and improve the performance of our systems. Our systems automatically categorized documents and identified relevant language in multi-million document corpora.Our data-driven approach to modeling and testing required corpus analysis techniques such as metadata analysis, corpus stratification, strategic sampling, concordance creation, text mining, and precision/recall metrics.Apart from on client projects, I contributed to the development and testing of internal toolsets, cross-functional communication efforts, and documentation of technical processes. Show less