Browse Prior Art Database

Accommodation Modeling for Online Social Networks Disclosure Number: IPCOM000235982D
Publication Date: 2014-Apr-01
Document File: 3 page(s) / 32K

Publishing Venue

The Prior Art Database


Disclosed is a system to perform automatic tagging and analysis of an online conversation to identify elements that indicate deviations in background, and, thus differences in usual language. The system then suggests to participants how to generate accommodations of deviations to reduce differences in language within the conversation.

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

Page 01 of 3

Accommodation Modeling for Online Social Networks

Online social networks (OSNs) are a universal mechanism to connect people and information in logical and organized ways, which enable sharing and processing of

information between the members. The most common mechanisms of sharing and processing information on online social networks are the wall, activity stream, timeline, or access to profiles. These mechanisms enable a user to rapidly share information with others and gather information from others in the networks.

Lexical accommodations in natural language enable a speaker to converge to a standard to appear more similar to each other. This regulation of communication includes speech convergence and accents.

For example, a common scenario of accommodation is:

1. Speaker A, a person from Boston, uses the word "wicked" in place of the

word "very"

2. Speaker B, a person from London, does not normally use the word "wicked"

  in the same context 3. Speaker B and Speaker A begin speaking 4. Speaker B asks, "Are you having a very good day, Speaker A?"

5. Speaker A says, "I am having a very good day."

6. Speaker A modifies the usual form of speech to converge to a standard such that "very" is used as the modifier instead of "wicked".

The OSNs lack the capabilities to generate lexical accommodations in speech and discussions, as person in a verbal or in-person conversation is able. In ONSs, the speed of conversation and textual nature of the conversation reduce the use of accommodations, and users miss the opportunity to build better relationships.

A system is needed to allow the use of accommodations in Online Social Networks

to generate the best online experiences.

The novel contribution is a system to optimize relationship generation in online social networks. The system identifies a conversation containing two or more users with backgrounds that deviate. The system then tags and analyzes the elements of the conversation that indicate deviations in background. Finally, the system suggests to participants how to generate accommodations of deviations.

The system can apply to two or more people, such as a group engaged in an online discussion. In addition, the system can resolve languages to regional or urban dialects or dictionaries (e.g., English dialects in the Southern United States vs. the Bronx, NY).

Further, the system can detect the severity of a conversation or subject and disable the optimizer in order to avoid inadvertently offending a user (e.g., add unintended humor about a person's accent). The invention may annotate messages indicating that an accommodation is taking place.

To implement the system in a preferred embodiment:

1. Determine deviation potentials

A. System extracts the user's identifiers in a conversation (e.g., email,


Page 02 of 3

user ide...