classification of rare recipes requires linguistic
play

Classification of Rare Recipes Requires Linguistic Features as - PowerPoint PPT Presentation

Classification of Rare Recipes Requires Linguistic Features as Special Ingredients Elham Mohammadi, Nada Naji, Louis Marceau, Marc Queudot, Eric Charton, Leila Kosseim, and Marie-Jean Meurs Banque Nationale du Canada Concordia University


  1. Classification of Rare Recipes Requires Linguistic Features as Special Ingredients Elham Mohammadi, Nada Naji, Louis Marceau, Marc Queudot, Eric Charton, Leila Kosseim, and Marie-Jean Meurs Banque Nationale du Canada Concordia University Université du Québec à Montréal

  2. Contents 2 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  3. Contents 3 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  4. Introduction 4 ❖ Motivation ➢ Many real-life scenarios involve the use of highly imbalanced datasets. ➢ Extraction of discriminative features Discriminative features can be used alongside ■ distributed representations.

  5. Introduction 5 ❖ Goal ➢ Investigating the efgectiveness of the use of discriminative features in a task with imbalanced data

  6. Contents 6 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  7. Dataset and Tasks 7 ❖ DEFT (Defj Fouille de Texte) 2013 (Grouin et al., 2013) ➢ A dataset of French cooking recipes labelled as ■ Task 1: Level of diffjculty Very Easy, Easy, Fairly Diffjcult, and Diffjcult ● ■ Task 2: Meal type Starter, Main Dish, and Dessert ●

  8. Dataset Statistics 8 Task 1 Task 2

  9. Contents 9 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  10. Methodology 10 ❖ Neural sub-model Embedding layer: pretrained BERT or CamemBERT ➢ Hidden layer: CNN or GRU ➢ Pooling layer: Attention, Average, Max ➢ ❖ Linguistic sub-model Feature extractor ➢ The extraction and selection of linguistic features was done ■ according to Charton et al. (2014) Fully-connected layer ➢

  11. Experiments 11 ❖ The joint model ❖ The independent neural-based sub-model ❖ Fine-tuned BERT and CamemBERT models

  12. Contents 12 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  13. Results: Task 1 13

  14. Results: Task 1 (Per-class F1) 14

  15. Results: Task 2 15

  16. Results: Task 2 (Per-class F1) 16

  17. Discussion 17 ❖ The joint model is more efgective in task 1 compared to task 2 The linguistic features used for task 2 ➢ might not be as representative of the classes as those for task 1 ■ are signifjcantly more sparse ■ The improvement caused by the joint model is higher ➢ in case of rare classes

  18. Contents 18 ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  19. Conclusion 19 ❖ In both tasks, the joint models outperform their neural counterparts ❖ The improvement by the joint models is higher in Task 1 ❖ The improvement by the joint models is more signifjcant for rare classes ❖ The strength of the joint architecture is in the handling of rare classes

  20. 20 Contents ❖ Introduction ❖ Dataset and Tasks ❖ Methodology ❖ Results and Discussion ❖ Conclusion

  21. Thank you!

Recommend


More recommend