Deep Network Solving Small-Big Data Issues in Sorting Emails
The Information age leveraged by new technology has made simple tasks like sorting emails, into all-consuming tasks. However, with efficient email analysis tools, important information can be made accessible without losing precious time while maintaining privacy and security. An email analysis solution considering different user and email aspects may be integrated optimally to rank importance of individual emails. The result leads to significant improvement of user’s experience for enhanced attention, organization, and sorting. Such analysis runs into the small big data problem due to the huge number of user analysis; nonetheless, each one of the users receives a limited size of data. A natural question arises on how to combine autonomous and user information; it is noteworthy that combining small and big data machine learning methods must become an essential part of the analysis. The above is done under smart features extraction and their calculation, while constructing the features is done according to user’s social network analysis, text analysis, etc. Furthermore, to enhance predictions at an individual level, cross-sectional learning is also included in the package. To be able to coordinate, the analysis is done as a network of models. Each model learns a different feature with information of the other models. The obtained network of models allows applying deep learning methodology in such case of “small-big data”.
Uri Itai is a machine learning researcher in Knowmail with strong knowledge in statistics and financial mathematics. He plays a key role within Knowmail’s Artificial Intelligence team, leading the cross data information research, namely trends of use and algorithmic development. His Machine Learning (ML) research focuses on a variety of areas, particularly deep learning (DL). His current work and experiences are based on both mathematical theory and practical programming. Prior to Knowmail he has extensive algorithmic research and was a data scientist in a cancer diagnostic startup. He received his PhD in Applied Mathematics in 2013 from The Technion – Israel Institute of Technology.