Browse Prior Art Database

System and methods for smart notification Disclosure Number: IPCOM000249088D
Publication Date: 2017-Feb-03
Document File: 5 page(s) / 131K

Publishing Venue

The Prior Art Database


System and methods for smart notification

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.


System and methods for smart notification

Intelligent detection of notification importance. Described system and methods allows users to determine if notification is important, or if it may be checked later. Natural language processing methods as well as cloud based solutions are used.

1. Client application installed on mobile devices disable notifications and fetch content of all notifications received by mobile device 2. Client application send notification content to server (cloud solution) 3. On server side notification content is used for semantic analysis and Named Entity recognition 4. Context/Tone of notification is fetched. Based on that as well as NER's importance of notification is set and send to mobile device 5. Based on importance and user's settings proper alarm is performed

Idea is to be implemented as client server application: 1) Client Mobile application installed on device. Provides simple UI to choose type of notifications that are to be taken into consideration (SMS, Emails, Slack notifications, Facebook messages etc.) and choose Predefined characteristics/types of these notifications (important, urgent, from_friends, from_family) or define new.

Example usage: 1. User installs application on mobile device 2. User chooses SMS and Emails notifications as notifications that he wants to be filtered 3. User chooses types of notifications: a) urgent b) spam c) family 4. User also wants to use sub types for family type: a) urgent b) may_wait 5. In next step user chooses settings how these notifications are to be alerted a) for urgent: vibro, signal, high volume b) for spam: no vibro, signal2, silent (no volume) c) for family: no vibro, signal3 d) for family urgent: vibro, high volume e) for family may_wait: low volume 6. User defines Named Entity and accepts Named Entity Suggested by application (that information is fetched from all users on server and pushed as suggestions for users) - "sale" goes to spam - "call back" goes to urgent - "Mom" goes to family

etc. 7. Application is ready to work in the background


Now every time users gets SMS or Email notification, application fetch its content and sends to server. As a response, application receives what type of notification should be used to alert user.

2) Server 1. Hosts types of notifications and based on Named Entities tries to generalized them to suggest users pre-defined types 2. Once notification content is received suitable ner algorithm implementation is used to fetch Name...