Anti-Spam Techniques Based on Artificial Immune System (Paperback)

Anti-Spam Technologies
Spam Problem
Prevalent Anti-Spam Technologies
Email Feature Extraction Approaches
Email Classification Techniques
Performance Evaluation and Standard Corpora
Summary

Artificial Immune System
Introduction
Biological Immune System
Artificial Immune System
Applications of AIS in Anti-Spam
Summary

Term Space Partition-Based Feature Construction Approach
Motivation
Principles of the TSP Approach
Implementation of the TSP Approach
Experiments
Summary

Immune Concentration-Based Feature Construction Approach
Introduction
Diversity of Detector Representation in AIS
Motivation of Concentration-Based Feature
Overview of Concentration-Based Feature
Gene Library Generation
Concentration Vector Construction
Relation to Other Methods
Complexity Analysis
Experimental Validation
Discussion
Summary

Local Concentration-Based Feature Extraction Approach
Introduction
Structure of Local Concentration Model
Term Selection and Detector Sets Generation
Construction of Local Concentration-Based Feature Vectors
Strategies for Defining Local Areas
Analysis of Local Concentration Model
Experimental Validation
Summary

Multi-Resolution Concentration-Based Feature Construction Approach
Introduction
Structure of Multi-Resolution Concentration Model
Multi-Resolution Concentration-Based Feature Construction Approach
Weighted Multi-Resolution Concentration-Based Feature Construction Approach
Experimental Validation
Summary

Adaptive Concentration Selection Model
Overview of Adaptive Concentration Selection Model
Setup of Gene Libraries
Construction of Feature Vectors Based on Immune Concentration
Implementation of Adaptive Concentration Selection Model
Experimental Validation
Summary

Variable Length Concentration-Based Feature Construction Method
Introduction
Structure of Variable Length Concentration Model
Experimental Parameters and Setup
Experimental Results on the VLC Approach
Discussion
Summary

Parameter Optimization of Concentration-Based Feature Construction Approaches
Introduction
Local Concentration-Based Feature Extraction Approach
Fireworks Algorithm
Parameter Optimization of Local Concentration Model for Spam Detection by Using Fireworks Algorithm
Experimental Validation
Summary

Immune Danger Theory-Based Ensemble Method
Introduction
Generating Signals
Classification Using Signals
Self-Trigger Process
Framework of DTE Model
Analysis of DTE Model
Filter Spam Using the DTE Model
Summary

Immune Danger Zone Principle-Based Dynamic Learning Method
Introduction
Global Learning and Local Learning
Necessity of Building Hybrid Models
Multi-Objective Learning Principles
Strategies for Combining Global Learning and Local Learning
Local Trade-Off between Capacity and Locality
Hybrid Model for Combining Models with Varied Locality
Relation to Multiple Classifier Combination
Validation of the Dynamic Learning Method
Summary

Immune-Based Dynamic Updating Algorithm
Introduction
Backgrounds of SVM and AIS
Principles of EM-Update and Sliding Window
Implementation of Algorithms
Filtering Spam Using the Dynamic Updating Algorithms
Discussion
Summary

AIS-Based Spam Filtering System and Implementation
Introduction
Framework of AIS-Based Spam Filtering Model
Postfix-Based Implementation
User Interests-Based Parameter Design
User Interaction
Test and Analysis
Summary