Browse Prior Art Database

Using device usage patterns and behaviours to create personalised content

IP.com Disclosure Number: IPCOM000245881D
Publication Date: 2016-Apr-15
Document File: 3 page(s) / 107K

Publishing Venue

The IP.com Prior Art Database

Abstract

This article describes a system that monitors a particular user's usage patterns when reading news articles and other documents. Using the gathered usage patterns, an article provider's information can be adjusted to be presented in a way that is optimal for the particular user. By producing altered articles tailored to a user's preferences, be they for simpler language, more images or a less opinionated article, providers can engage a user to a much greater extent and thus enabling a greater commercial viability from repeat business and more effective advertising placement.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Using device usage patterns and behaviours to create personalised content

Current content on the internet, such as news stories or articles, are rarely personalised to individual reading and visualisation preferences. Some systems, such as Google News, recommend existing content based on the user's interests by looking at browsing history. However in these systems the content is not altered or personalised and the user only gets a hyperlink to the original source. This disclosed system goes a step further by analysing the user's device usage patterns and re-structuring the existing content to satisfy the reading and visualisation preferences of the user. Herein disclosed is a solution to track the user's device usage patterns using a variety of methods when viewing content on device, and use this data to determine how the user prefers to read and visualise information. With this data, and subject requests by the user, the system generates personalised content of the requested subject based on the unique preferences of the user, including subsets of different articles based on enjoyed style, complexity, images and other similar metrics.

    The disclosed system will first track usage patterns on the user's devices. Methods to do this include, but are not limited to, using a front facing camera to track engagement, expressions and concentration, tracking how fast a user scrolls though news articles and tracking how long a user takes to read articles. Following this stage, in order to analyse the data to determine the user's reading patterns, the system first identifies categories in the content such as quotes, images, headings and text in the content being visualised and classify them into the pre-determined 'data type' categories. The system also differentiates between and classifies facts and opinions by using semantic analysis and flagging harsh language. Whilst the user is visualising the content, the system tracks how long the user spends on each category, which categories get more attention from the user, and therefore which type of data they prefer for the type of content they are visualising.

Below is an example of such a data gathering stage:


Original article:
Football team wins match

Today saw a last-minute victory by X United, beating their rivals Y City 2-1. Top-scoring centre forward John Football had this to say: "It was a great game, obviously I'm happy to get the goal but the main thing is we got the three points".

This superlative, titanic performance by X United was epitomised by their pass completion rate of 87% and their total run distance of almost 1km per player more than the opposite team.

Analysis of the article with tags:


Football team wins match

Today saw a last-minute victory by X United, beating their rivals Y City 2-1.

Top-scoring centre forward John Football had this to say: "It was a great game, obviously I'm happy to get the goal but the main thing is we got the three points".

This superlative, unbelievab...