Communication via E-Mail is still one of the biggest interaction channels worldwide, and will most likely continue to be one in the next years. Therefore, augmenting this channel is something helpful for e.g. Service Desks, Sales Teams or Recruiting Teams. One way to do so is via detecting the type of attachment an E-Mail has. For instance, mails that are sent to HR for potential hiring may have six to eight typical attachment types, one of it being CVs.
By detecting the type of content, automated routings can be developed. Mails with CV attachments can be integrated with the inhouse HR application, such that new entries are automatically being synced. Together with an Information Extraction approach, CVs can also be parsed and the then structured data can be captured by the given system.
Of course, such models require plenty of training data, which is mostly hard to acquire due to the lack of labeling resources. Additionally, such data is mostly sensitive, such that you don’t want to outsource it. With refinery, you can remove this bottleneck easily. Our system is designed to label data on large-scale within hours instead of weeks, and helps you to maintain a high data quality.
Are you interested? refinery is open-source and free to use, so give it a try - we’ll support you along the way at any time you need guidance or help.