Browse Prior Art Database

Contextual Social Content Harmonization Disclosure Number: IPCOM000247224D
Publication Date: 2016-Aug-17
Document File: 3 page(s) / 27K

Publishing Venue

The Prior Art Database


Disclosed is a method of performing contextual harmonization of social media content and sentiment to allow for a more accurate visual comparison of sentiment across users, based on the user’s psycholinguistics profile of learned personality attributes and semantic analysis of content.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 46% of the total text.

Page 01 of 3

Contextual Social Content Harmonization

Data harmonization is the improvement of data quality and utilization through machine learning capabilities. Data harmonization interprets existing characteristics of data and action taken on data, and then uses that information to transform or suggest subsequent data quality improvements. Harmonization creates the possibility of combining data from heterogeneous sources into integrated, consistent, and unambiguous information products, in a way that is of no concern to the end-user.

For example, a single concept can be represented in three different ways in three different sources (e.g., USA, U.S.A, and United States of America). Although each representation has the same meaning, data standardization is required for data analysis. Different rules are applied for data standardization.

In the case of user-generated data, the challenges are greater; therefore, contextual harmonization is required. Following are examples of the need for contextual harmonization of social content:

Example 1:

"People in the United States and Northern Europemay prefer direct communications, while people in other countries may prefer indirect communications. When denying a request in the U.S., a writer will typically apologize, but firmly state that request was denied. In other countries, that directness may seem rude. A writer may instead write that the decision has not yet been made, delaying the answer with the expectation that the requester will not ask again. In other countries, this is viewed as more polite than flatly denying someone; however, in the United States this may give false hope to the requester, and the requester may ask again."*

Example 2:

Depending on a person's beliefs, the contextual content can have different meanings. For example, some people are very conservative (e.g., User A) about entering any positive comments, and some people are very liberal (e.g., User B) about entering positive comments. The sentiment for social communications may both provide positive comments; however, each might have different meanings. Thus, the consolidated degree of positive feedback is not accurately identifiable. In this example, if User A provides feedback, "OK", then it is actually "good", and if User B provides the feedback, "Very good", then it actually, "good".

A method is needed to perform an ongoing analysis of a person's personality based on social content with the capability to perform contextual harmonization of social content sentiment across user types.

The novel contribution is a method for performing contextual harmonization of social media content and sentiment to enable accurate visual comparisons of sentiment across users, based on the user's psycholinguistics profile of learned personality attributes and semantic analysis of content. By understanding a user's personality


Page 02 of 3

attributes, the system can reconcile or map the user's sentiment on a harmonization scale so that compari...