Aggregate Fraud Score Dynamic Scoring Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting online fraud, such as phishing and identity theft, are inadequate as fraudsters continually develop countermeasures, including frequent IP address changes to evade blacklisting, necessitating the development of new fraud-identifying tests to keep pace with evolving fraud techniques.
Innovation Solution
A method employing a computer system to determine an aggregate fraud score by combining multiple fraud scores using distinct evaluation procedures, with a third score modifying the aggregate score based on a product of the difference between the aggregate score and a maximum allowable score, to dynamically assess the legitimacy of online documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple fraud scores are combined to improve detection accuracy, then the reliability of fraud detection is improved, but the device complexity increases
Solution Approach 1:
The fraud detection system is segmented into multiple independent fraud-evaluation procedures, each producing a separate fraud score. These segmented evaluation procedures assess different aspects of document legitimacy independently, then their scores are combined to form an aggregate fraud score. This segmentation allows the system to improve detection reliability through multiple perspectives while keeping each individual evaluation module relatively simple and manageable.
Solution Approach 2:
Multiple fraud scores from distinct evaluation procedures are merged into a single aggregate fraud score. The system combines the outputs of various fraud-evaluation procedures using a defined aggregation method, allowing the benefits of multiple detection approaches to be consolidated into one comprehensive fraud assessment, thereby improving overall detection accuracy without requiring complex integration logic.
2Adaptability or versatility
If new fraud-identifying tests are continuously developed to counter evolving fraud techniques, then the adaptability of the detection system is improved, but the device complexity increases
Solution Approach 1:
The fraud detection system is designed to dynamically incorporate new fraud-identifying tests and evaluation procedures. The aggregate fraud score is computed by combining scores from multiple evaluation procedures, allowing new tests to be added to the system without requiring complete redesign. This dynamic structure enables the system to adapt to evolving fraud techniques while maintaining a manageable level of complexity through modular integration.
Solution Approach 2:
The fraud detection system employs multiple universal fraud-evaluation procedures that can assess various types of fraudulent documents. Each evaluation procedure is designed to be broadly applicable to different fraud scenarios, allowing the system to handle diverse fraud techniques with a unified framework. This multi-functionality approach enables the system to adapt to new fraud types without requiring entirely new specialized mechanisms for each scenario.
3Measurement precision
If fraud scores are continuously updated to reflect new fraud patterns, then the measurement precision of fraud detection is improved, but the loss of time increases
Solution Approach 1:
The system performs preliminary fraud evaluation by computing individual fraud scores from multiple evaluation procedures before aggregating them into a final fraud score. This preliminary action allows new fraud patterns to be detected and scored in advance, with results ready for rapid aggregation when needed. The pre-computed scores can be quickly combined to produce updated aggregate scores, reducing the time required for continuous fraud assessment while maintaining measurement precision.
Solution Approach 2:
The fraud detection system efficiently processes fraud scoring by skipping redundant evaluation steps. When new fraud patterns are identified, the system updates only the specific evaluation procedures affected rather than re-running all evaluations from scratch. This selective updating approach allows rapid score computation and aggregation, maintaining high measurement precision while minimizing the time loss associated with continuous fraud score updates.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
In some embodiments, an online fraud prevention system combines the output of several distinct fraud filters, to produce an aggregate score indicative of the likelihood that a surveyed target document (e.g. webpage, email) is fraudulent. Newly implemented fraud filters can be incorporated and ageing fraud filters can be phased out without the need to recalculate individual scores or to renormalize the aggregate fraud score. Every time the output of an individual filter is calculated, the aggregate score is updated in a manner which ensures the aggregate score remains within predetermined bounds defined by a minimum allowable score and a maximum allowable score (e.g., 0 to 100).