|
Work item:
|
X.tc-ifd
|
|
Subject/title:
|
Technical capabilities of interactive fraud detection
|
|
Status:
|
Under study
|
|
Approval process:
|
TAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2024-09 (Low priority)
|
|
Liaison:
|
-
|
|
Supporting members:
|
-
|
|
Summary:
|
Fraud detection is a very important capability for certain ICT systems including systems for payment, rental system and credit granting etc. Traditional fraud detection mechanisms for these systems are based on service information gathering and rule/pattern matching etc. As the big data and AI technologies evolving, system based automatic decision-making fraud detection are more and more common. But the fraud identification decisions made only by back-end systems will have some challenges such as the dependency on adequate data, balance between accuracy and recall rate and a friendly user interaction when service abortion caused by fraud detection and response. Interactive fraud detection, which means during the service launching and execution, users are invited in a conversation with ICT service systems, for information query to identify and reduce fraud risks. Some of the benefits of interactive fraud detection are:
Firstly, interactive fraud detection will help to gather necessary information during the service process to strengthen the detection and decision of fraud analysis, which can improve the accuracy and reduced the recall rate.
Secondly, when there is a suspicious fraud happened, an interaction with user will help to let the user understand what happened and potentially promote the user’s fraud risk awareness in subsequent transactions.
Considering these factors, this contribution intends to introduce technical capabilities of interactive fraud detection, with which, the fraud detection module of an ICT system will cooperate with an interactive engine to identify the fraud risk, when there is no enough information to decide the risk level. According to rules, corpus and some other situation, interactive engine contact users by AI dial-out, text-alert or some other way to ensure the risk level to risk identification engine.
This Recommendation specifies the technical capabilities of interactive fraud detection, including the capability identification, architectural framework to support the identified capabilities and procedures that support the interaction among different architectural components.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2022-05-23 20:47:36
|
|
Last update:
2024-06-11 14:40:06
|
|
