Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Emotion analysis in live chat sessions

IP.com Disclosure Number: IPCOM000243071D
Publication Date: 2015-Sep-11
Document File: 3 page(s) / 155K

Publishing Venue

The IP.com Prior Art Database


A system and method for determining a person's emotional mood while having a chat is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 3

Emotion analysis in live chat sessions

Disclosed is a system and method for determining a person's emotional mood while having a chat. The system may provide options for logging the conversations and opportunities for statistical analysis of a group of conversations. In an embodiment, the system analyzes the use of adjectives that characterize the nouns in a sentence and based on the results provides a visual notification of the mood on the chat window screen. The system identifies nouns and for adjectives that describe those nouns. A

database containing a dictionary is used. The dictionary provides information that is used by a scoring system that can be generated based on this analysis. Based on this scoring system changes may be made to the chat window showing live updates to the mood setting. Additionally the chat session scoring may be recorded in another result

storage database (RSD). The RSD may be used for future statistical analysis.

Figure 1 depicts a flow of an emotional analysis system (EAS) monitoring a chat conversation.

Figure 1

Processing of text in the chat session may include the following considerations:

Expect that a lot of the texts are incomplete sentences with slang and internet lingo. Not all conversations are emotional; however, each blurb in chats tends to be short and more expressive.

The analysis uses the following guidelines:

Search for expression angularities that determines strong emotions. Example: LOL,


dammit, hmm, ouch, etc.

Search for interjections; if there is one, look for the relationship of word with that

expression (such as !, !!!, …, ?, etc).


Page 02 of 3

Some textual parsing, such as negation, use of adjectives, and changing of words in


a conversation (one person may say 'he' and the other side says 'dufus') Look up all the found wor...