Traditional data companies do not support annotation of the data in other languages, excepting Chinese and English. Iceberg Data has full-time annotators of more than 300 native languages, so that your multilingual training data can be quickly obtained
Iceberg Data owns an industry-leading data processing platform - Iceberg Data annotation system. This system is designed by senior designer of the industry. And this system integrates into data annotation tools and project process management to protect data delivery.
automatic and accurate bidding
strictly selected annotators
six guarantee systems
deeply developed quality control
intelligent assist of algorithm
saved project cost
Operation of Iceberg Data annotation system
"High-level data quality is very important to our intelligent driving project, or, there will be irreparable algorithm potential safety hazards. We have used multiple internal tools to evaluate the annotation data delivered by Iceberg Data team. The results of quality evaluation are good. And we are confident to apply these data into important algorithm training projects."
———— Mr. Qin
the responsible person of Baidu intelligent driving project
Three million Q&As have been extracted from 100,000 articles for the customers for deep learning of NLP reading comprehension.
Recording audios of customer service of 30,000 hours are listened to distribute annotation and help customers' algorithm training for intelligent customer service system.
Two million of current news are classified, and labels corresponding to the contents have been made to help the customers to develop personalized recommended algorithm products.
Iceberg Data is owned by Yu'an (Beijing) Technology Co., Ltd. It specializes in providing quick and efficient basic data service for AI industry. By using independently researched and developed Iceberg Data annotation system and combining with deeply developed global crowdsourcing and self-operating factory mode, Iceberg Data team is able to provide customized data processing scheme and implementation service in accordance with specific domain demands and scenarios to deliver standardized and structured high-precision machine learning data.