Resources Contact Us Home
Spam filtering using feature relevance assignment in neural networks

Image Number 11 for United States Patent #8131655.

In some embodiments, a spam filtering method includes computing a pattern relevance for each of a set of message feature patterns, and using a neural network filter to classify incoming messages as spam or ham according to the pattern relevancies. Each message feature pattern is characterized by the simultaneous presence within a message of a specific set of message features (e.g., the presence of certain keywords within the message body, various message header heuristics, various message layout features, etc.). Each message feature may be spam- or ham-identifying, and may receive a tunable feature relevance weight from an external source (e.g. data file and/or human operator). The external feature relevance weights modulate the set of neuronal weights calculated through a training process of the neural network.

  Recently Added Patents
Biaxially oriented film which could be thermally laminated with paper and other substrates
Touchscreen with extended conductive pattern
Method and apparatus for representing sensory effects and computer readable recording medium storing user sensory preference metadata
Method and apparatus for content-aware resizing of data chunks for replication
Horizontal aggregations in a relational database management system
MOS transistor having combined-source structure with low power consumption and method for fabricating the same
Catalysts for hydrodeoxygenation of polyols
  Randomly Featured Patents
Valve mechanism
Shower arrangement
Method and apparatus for obtaining and using event data recorder triage data
Face driver
Seating-system for folding wheelchairs
Fast sense amplifier for small voltage differences
Continuous flow expandable chamber and dynamic displacement rotary devices
Liquid crystalline bis-chlorocarbonic acid esters
Lens covering and cleaning apparatus
Modulated high frequency dimmer circuit with infrared suppression