Browse Prior Art Database

Method and apparatus for semantic driven screen touching point calibration

IP.com Disclosure Number: IPCOM000232558D
Publication Date: 2013-Nov-15
Document File: 5 page(s) / 73K

Publishing Venue

The IP.com Prior Art Database

Abstract

In many languages (like Chinese), users usually will change/insert/delete based on a word granularity (that is sematntically reasonable) instead of on a single character level. So we are motivated to facilitate such behavior by automatically help users refocus/calibrate their inaccurate touching into the semantic meaningful areas. The invention provides a novel method & apparatus to automatically determine the suitable touch point (cursor position) for the users when the user's touching is sometimes ambiguous and not accurate physically.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 01 of 5

Method and apparatus for semantic driven screen touching point calibration

The problem of recalibration of the inaccurate and noisy user input on the touch screen is solved byusing the word segmentation prior information. This is particularly important in the Character language like Chinese , Korean, etc. Current touch screen input interface has never considered this particular finding and no technology is presented to solve this specific problem .

The sementic and word level information is explored for the 'going-to-manipulate' sentence, this feature helps calibrate the inaccurate and nosiy user input (the touching location on the screen ) can be recalibrated to the actual location . This improves the user experience and is particularly important for the old people or people who are not familiar with touch screen , especially when the device is very small.

1



Page 02 of 5

Core idea illustration using Chinese sentence as an example :

2



Page 03 of 5

3



Page 04 of 5

4



Page 05 of 5

In summary, this disclosure proposes a method for detecting and rectifying user 's touching input


1) detecting the touch location nearby semantic structure


2) detecting the touch location by historical behavior induced bia compensation


3) infering the user profile thru touching behavior

5