|
Work item:
|
PSTR.CEC (ex TR-CEC)
|
|
Subject/title:
|
Framework for Quality of Emotional Experience (QoEE) by using cultural and emotional context in applications based on natural language processing
|
|
Status:
|
Agreed on 2026-06-17
|
|
Approval process:
|
Agreement
|
|
Type of work item:
|
Technical report
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
-
|
|
Liaison:
|
ITU-T SG13, ETSI TC STQ
|
|
Supporting members:
|
India
|
|
Summary:
|
Natural language processing (NLP) is being employed to support human-machine interaction for effective consumer engagement. NLP-based applications employ both probabilistic and non-probabilistic methods to gauge user intent and thereafter render appropriate responses. However, the accuracy of gauging the user intent depends on factors including the learning model, training corpus, clarity of the input and scope of the domain.
Contextual information such as emotional and cultural inputs can be utilised for increasing effectiveness of NLP applications such as chatbots. Since language often varies across cultures and may convey different meanings and nuances depending on the context and an individual’s emotional state can influence communication style, word choice, and behaviour, incorporating such contextual information in NLP-based applications can support more personalized and “context-aware” interactions.
Today, most NLP-based applications including large language models (LLMs) endeavour to utilise and extract additional context to make them “context-aware”. Future chatbots are expected to become increasingly capable of adapting to user preferences and corresponding emotional nuances. Such capabilities may improve the quality of interactions and contribute to a more positive user experience. In this context, the concept of quality of emotional experience (QoEE) can be considered as “the effect that the application has on the deeper feelings of an individual”. It is thus an essential subset of quality of experience (QoE), which is defined as the “degree of delight or annoyance of the user of an application or service”.
This Technical Report examines methods for the objective assessment of QoEE in text-based chatbot interactions and provides key performance indicator (KPI) metrics that may support the evaluation of QoEE in future applications.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2023-02-06 09:35:47
|
|
Last update:
2026-06-24 09:54:57
|
