Browse Prior Art Database

Method and System for Clustering Messages from a Digital Communication Service in a User Device

IP.com Disclosure Number: IPCOM000241881D
Publication Date: 2015-Jun-05
Document File: 2 page(s) / 35K

Publishing Venue

The IP.com Prior Art Database

Related People

Rick Banerjee: INVENTOR [+2]

Abstract

A method and system is disclosed for displaying one or more messages from a digital communication service in one or more clusters in a user device. The digital communication services are, but not limited to, email and Short Message Service (SMS). The user device is, but not limited to, a mobile phone and a tablet.

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 52% of the total text.

Method and System for Clustering Messages from a Digital Communication Service in a User Device

Abstract

A method and system is disclosed for displaying one or more messages from a digital communication service in one or more clusters in a user device.  The digital communication services are, but not limited to, email and Short Message Service (SMS).  The user device is, but not limited to, a mobile phone and a tablet. 

Description

Disclosed is a method and system for displaying one or more messages from a digital communication service in one or more clusters in a user device.  The digital communication services are, but not limited to email and Short Message Service (SMS).  The user device is, but not limited to a mobile phone and a tablet. 

The method and system creates clusters for the one or more messages in a digital communication service by processing text in the one or more messages.  The one or more clusters contain the one or more messages that relate to a particular topic.

The method and system performs feature selection for removing redundant words and feature transformation for reducing dimensionality of the one or more messages before clustering.  The method and system performs feature selection based on one or more well-known techniques.  The techniques are, but not limited to, document frequency based, term strength, term contribution and entropy reduction.  The method and system performs feature transformation using one or more techniques that are, but not limited to, latent semantic indexing.  Thereafter, the method and system performs the clustering.  The method and system uses one or more text clustering techniques for clustering the one or more messages.  The text clustering techniques are, but not limited to, online text clustering technique, hierarchical clustering methods, k-means algorithm, scatter/gather method and fractionation.

The digital communication services are classified into two categories based on the manner in which the one or more messages are stored in the services.  The categories are:

Category A: The services in which one or more messages are stored by the service provider such as email; and

Category B: The services in which the one or more messages are stored on the user device such as SMS.

For category A, the method and system processes the one or more messages stored in a server of the service provider for clustering the one or more messages.  

The ...