Avaliação de resultados

25 outubro 2017, 11:30 • Pável Pereira Calado

- Exercícios de avaliação (precisão, recall, etc.)

Information Extraction

20 outubro 2017, 15:30 • Bruno Emanuel Da Graça Martins

• Information extraction with nltk
  
• Exercises about regular expressions
  
• Exercises about Hidden Markov Models
  
• Support to the course project
  

Information Extraction

20 outubro 2017, 15:30 • João Miguel Cordeiro Monteiro

• Information extraction with NLTK
  
• Regular expressions
  
• Hidden Markov Models
  
• Support to the course project
  

Evaluation of IR and IE Systems

20 outubro 2017, 12:30 • Bruno Emanuel Da Graça Martins

• Why and what to evaluate?
• The Cleverdon evaluation methodology
• Measuring precision and recall in terms of retrieving relevant documents
• Precision-recall curves and interpolated curves
• Other evaluation metrics:
• P@N and R-precision
• F1-metric
• Mean Average Precision (MAP)
• Normalized Discounted Cumulative Gain (NDCG)
• Mean Reciprocal Rank (MRR)
• Metrics for comparing Rankins (Spearman's Rho and Kendall's Tau)
• Methods for collecting reference data
  

Evaluation of IR and IE Systems

20 outubro 2017, 12:30 • Bruno Emanuel Da Graça Martins

• Why and what to evaluate?
• The Cleverdon evaluation methodology
• Measuring precision and recall in terms of retrieving relevant documents
• Precision-recall curves and interpolated curves
• Other evaluation metrics:
• P@N and R-precision
• F1-metric
• Mean Average Precision (MAP)
• Normalized Discounted Cumulative Gain (NDCG)
• Mean Reciprocal Rank (MRR)
• Metrics for comparing Rankins (Spearman's Rho and Kendall's Tau)
• Methods for collecting reference data