Browse Prior Art Database

Augmented Skills based Content Translation

IP.com Disclosure Number: IPCOM000254239D
Publication Date: 2018-Jun-13
Document File: 3 page(s) / 217K

Publishing Venue

The IP.com Prior Art Database

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

Augmented Skills based Content Translation A presenter/content writer needs to know the content well enough to customize it to a specific audience/skill level. An instructor may have to choose different materials based on the skill level of the students. Currently, there is no way to effectively dynamically alter content based on understanding the skill level of the consumer compared to the skill level of the original content. The novel contribution is a system that can develop an understanding of a reader’s skill level and then translate the original content to a more appropriate level that the reader can easily understand. This system enables the reader to learn based on current needs and abilities. For example, homework instructions could dynamically change based on the skill level of the individual student. Implementing this system in a preferred embodiment entails the following:

1. Establishing the baseline: system parses the original content to determine the skill level of the intended audience (e.g., beginner, intermediate, advanced); system can use Natural Language Processing (NLP) for this step

2. Evaluating the audience: reader engages with content and the system determines the associated baseline skill level

A. Skill level obtained via scans - how many times has the audience re-read the same passage

B. User feedback questionnaire: dynamically summarize the content C. Historical comparison of interaction with content: derive the historical

baseline per user versus the per group D. Social network data (e.g., job history on a professional network website,

skills identified by the user, etc.) E. Shared content and user behaviors (e.g., user shares articles related to

machine learning, so likely not a novice. User reads advanced websites related to mobile development, likely not a novice. User has never read an article about Digital DSLR cameras, likely a novice)

3. Capturing and storing the baselines in a reusable repository or corpus 4. Skill Matching: Does the skill level match from Step 1 and Step 2?

A. If yes, then the system takes no action B. If no, then the system translates the content to the appropriate level for the

reader; establishes the comparison with the baseline (i.e., what is the delta)

5. Taking Action: translate the content using the data collected from Steps 1 - 4 A. Content could be originally created by the content creator where it is

marked as novice/beginning, intermediate, and advanced either manually or through machine learning analyzing each section of the content

B. Content could be completely generated through artificial intelligence (AI). For example, a content creator might write just the more advanced versions and beginning cont...