A Time-Sensitive Spam Filter Algorithm Dealing with Concept-drift
Jiaolong Liu · Advances in engineering research/Advances in Engineering Research · 2016
Spam, under a variety of shapes and forms, continues to inflict increased damage.Varying machine learning techniques have played an important role in spam filtering field in condition that ample training data is available to build a robust classifier.These methods include Decision Tree, Support Vector Machine (SVM), etc.However, spam filtering is a particularly challenging task as the data distribution and concept being learned changes over time.More seriously, data stream classification poses many challenges to the data mining community.In this paper, we proposed a time-sensitive spam filter algorithm dealing with concept-drift (TSSFA), which is appropriate for such dynamically changing contexts.We evaluate its performance on the TREC public corpus, and showed satisfactory result.