Microsoft Technology Licensing, LLC

Yi Luo

Mainak Sen

Ravi Kiran Reddy Poluri

Sharada Shirish Acharya

Christian Rudnick

Weigsheng Li

Nan Li

Jing Luo

An Yan

Emails or other communications are labeled with a category label such as 'spam' or “good” without using confidential or Personally Identifiable Information (PII). The category label is based on features of the emails such as metadata that do not contain PII. Graphs of inferred relationships between email features and category labels are used to assign labels to emails and to features of the emails. The labeled emails are used as a training dataset for training a machine learning model (“MLM”). The MLM identifies unwanted emails such as spam, bulk email, phishing email, and emails that contain malware.

Privacy-preserving labeling and classification of email

Emails or other communications are labeled with a category label such as 'spam' or 'good' without using confidential or Personally Identifiable Information (PII). The category label is based on features of the emails such as metadata that do not contain PII. Graphs of inferred relationships between email features and category labels are used to assign labels to emails and to features of the emails. The labeled emails are used as a training dataset for training a machine learning model (&#x201c;MLM&#x201d;). The MLM model identifies unwanted emails such as spam, bulk email, phishing email, and emails that contain malware.

Methods are provided for improving the ability to apply modeling techniques similar to relational algebra to an expanded number of workflows. By allowing a relational algebra type modeling technique to be applied to an expanded number of workflows, an increased number of data processing workflows can be more readily improved, such as by automatic modification of the sequence of tasks in a workflow, to reduce the execution costs for a workflow. The relational algebra type modeling technique can also allow for identification of portions of data processing workflows or queries that share a common input and output.

Extending relational algebra for data management

Growing community of inventors

Seattle, WA, United States of America

Yi Luo