Mecurate: Personalized curation service using a tiny text intelligence

The necessity and importance of intelligent content curation have been demanding, as exemplified by Google Photos and Inbox. In this demo, we demonstrate a personalized curation service on a smartphone, called meCurate. By understanding implicit user interests from user smartphone usage data, meCurate intelligently organizes both in-device user content (e.g., SMSs and bookmarks) and external content (e.g., news articles crawled from The New York Times website), which are likely to match the inferred user's interests. In addition, meCurate retrieves semantically relevant in-device user content and external content to an explicit text or voice query. To this end, we utilize a tiny text intelligence which is suitable for smartphones with limited resources. Notably, meCurate works in a stand-alone, privacy-protecting manner without sending out any in-device personal data or content, resulting in a unique user experience.