Ontology Linguist, AI Products

About DoubleVerify DoubleVerify is the premier independent provider of marketing measurement software, data, and analytics. It verifies the quality and effectiveness of digital media for the globe's most extensive brands and media platforms. DV ensures media transparency and accountability, delivering the highest quality impressions for optimal advertising performance. Since 2008, DV has assisted hundreds of Fortune 500 companies to get the most from their media spend. They have achieved this by offering top-tier solutions across the digital ecosystem, which contribute to a better industry. Learn more at www.doubleverify.com. Position Overview DoubleVerify employs an automatic classification engine to classify millions of webpages each day. This ensures advertisers can only place their ads next to content that aligns with their brand's suitability preferences and avoid dubious content. This role is essential in supporting the automatic classification engine. In this role, you will develop a multilingual ontology used for automatic text classification. This is done by finding key phrases and contextual conditions to facilitate the automatic classification system's recognition of topics in texts. Moreover, you will be responsible for improving the mapping of these ontology topics to classifications by developing mapping conditions. Throughout your work, you will use statistical metrics to enhance overall accuracy. This role does not offer support for relocation or work permit processes, and it will be a hybrid position with 1-2 days per week in the office. Essential Skills The ideal candidate must be fluent in both spoken and written English, and they should have Japanese language skills (fluent or professional level). They must have strong teamwork and communication skills, be attentive to detail, and have critical thinking and problem-solving abilities. The ability to prioritize and organize workload, as well as computer literacy, are also essential. Preferred Skills It would be beneficial for the candidate to have the ability to use statistics in decision-making and to have relevant education, like linguistics, languages, or language technology. Nice to have but not required Prior knowledge of ontologies, understanding of regular expressions, and knowledge of the git version control system would be advantageous but are not mandatory. Not-so-fun fact: Research shows that men apply for jobs when they meet approximately 60% of the job criteria. In contrast, women and other marginalized groups usually only apply when they fulfill every requirement. So, if you think you have what it takes, but you're not sure you check every box, feel free to apply anyway!