🏷 Text classification

=== the process of categorizing elements into organized groups==

Formal definition

• Input:
  • a pattern: $d$ (?)
  • fixed set of classes: $C=\{c1,c2,cj\}$
• Output:
  • predicted class: $c\in C$

Methods

• 🌙 Symbolic (deep) processing:
  • → hand coded rules
• 👮‍♂️ Supervised machine learning:
  • → 🏷 ML Classifiers

Multiple classes

• Build a classifier for each class
• Compare the results of each classifier and chose the highest result

Evaluation

Measures

• ⚖ Balanced F measure
  • 🏹 Precision
  • 🛒 Recall
• 🎯 Accuracy

Averaging

• Macroaveraging: Compute performance for each class, then average
• Microaveraging: Collect decision for all classes, compute contingency table, evaluate

📖 Example:

Spam or not?