Collaborative Email Filtering via Shared Filter Expressions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current email filtering systems are inadequate in effectively reducing unsolicited emails, leading to decreased productivity and wasted network bandwidth, especially on mobile devices, as spam messages often bypass existing filters and require manual removal.
Innovation Solution
A collaborative spam filtering system where devices register with a remote spam filter service to receive updates for specific filter expressions, allowing for real-time identification and filtering of spam messages based on shared criteria, such as originating addresses or content, and applying these expressions to incoming emails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional spam filters are implemented on messaging servers, then some spam messages are blocked, but many spam messages still get through and require manual removal
Solution Approach 1:
The patent combines multiple spam filtering systems into a collaborative network where each device contributes to and benefits from a shared filtering capability. Filter expressions are generated locally on multiple devices and then shared across the network, creating a collective filtering system that is more effective than individual filters while reducing manual intervention needs.
Solution Approach 2:
The system implements feedback mechanisms where filter expressions are continuously generated, shared, and updated across the collaborative network. When new spam patterns are identified by any device, the resulting filter expressions are propagated to other devices, creating a dynamic feedback loop that continuously improves filtering effectiveness without requiring manual reconfiguration.
2Reliability
If individual devices maintain their own spam filters, then each device can independently filter emails, but the overall filtering effectiveness is limited and redundant
Solution Approach 1:
The patent creates a universal filtering system where a single filter expression generated on any device can be applied across the entire collaborative network. This multi-functional approach allows the same filter logic to serve multiple devices simultaneously, improving reliability through shared intelligence while reducing the complexity of managing separate filters on each device.
Solution Approach 2:
Filter expressions are copied and propagated across the collaborative network rather than being independently created on each device. This copying mechanism allows proven filter logic to be replicated and applied universally, improving reliability through consistency while reducing the complexity of original filter development on each device.
3Productivity
If all filter updates are received by each device, then comprehensive spam coverage is achieved, but unnecessary updates consume device resources and bandwidth
Solution Approach 1:
The patent applies local quality by allowing each device to selectively receive and apply filter expressions based on its specific needs and characteristics. Rather than uniformly distributing all filter updates to all devices, the system enables targeted distribution where filter expressions are applied where most effective, improving spam detection coverage while reducing unnecessary processing energy consumption on devices where the filters are less relevant.
Data Source
Figure 1
Figure 2
AI summary
A method is provided for collaboratively maintaining a filter program for a device. The method comprises the following steps. The filter program is registered with a remote program in order to receive spam e-mail filter updates. Notification messages are then received from the remote program. The notification messages comprise spam e-mail filter updates defining at least one filter expression received at the remote program from a third party. The filter expression is applied at the filter program for identifying spam e-mail messages and the spam e-mail messages to the device are filtered. A computer readable medium comprising instructions for executing the method and a computing device configured to implement the method are also provided.