Within SAP HANA, text analysis provides access to unstructured textual information for linguistically aware indexing and searching, semantic searching and transformation of unstructured into structured information for analytics and correlation with structured data sets. Via Natural Language Processing, it automatically extracts entities for analysis.
• Built an industry-leading German sentiment analysis module for social media
• Data curation and annotation for named-entity recognition in German, Dutch, Spanish, English
• Regression testing to improve precision and recall for sentiment and entity extraction
• Maintained remote build machine staging area
• Used statistical finite-state parsers, part-of-speech taggers, stemmers, tokenizers
• Corporate training for Agile development and software best engineering best practices