Portfolio Category: Data

Lead the launch of accounts recommender / similar companies (2017)  using  domain specific word2vec (CNN)  and model features extracted from corporate descriptions, and news articles including: Company specialities/ offerings (harvested via descriptions) Company popularity Location Revenue Employee count News agents count News count Co-occurence of companies in 'email watch list' Co-occurence of companies in news…

Launched contact matching service @94.5% (+17%) accuracy and @89.8% (+23%) recall to enable CRM contact "Refresh" product launch in Q3'2018. Record linking model features include: Executive name Employment titles Skill based similarity Job level, job function based similarity Title co-occurrence based similarity Email match Canonical LI URL Executive biography Education Age Location Other social handles

Increase firmographic richness (completion) using web crawls, machine assisted modules,  and human-in-the-loop pipelines for top 1M companies to 90% coverage across all key attributes (2018):  addresses, industry classifications, revenue, employee counts, and URLs.  Lead to industry leading company data quality @90% accuracy. Examples of breadth and depth of primary sources includes: US Department of Labor…

Scoped, provided outside in knowledge through partnerships / OEM vendors, and developed monetization strategy for Entity matching / record linkage as-a-service.  Migrated from rule based record linkage to ML based models (random forest & logistic regression) using hadoop and spark applications with 20% increase in F1 and 40% increase in recall while maintaining 90% precision. …