zpwiki/pages/topics/question/README.md

90 lines
2.2 KiB
Markdown
Raw Normal View History

2020-03-06 09:53:42 +00:00
# Question Answering
2020-06-11 10:07:46 +00:00
Task definition:
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
- Create a clone of SQuaD 2.0 in Slovak language
- Setup annotation infrastructure
- Perform and evaluate annotations
- Consider using machine translation
- Train and evaluate Question Answering model
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
## Tasks
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
### Raw Data Preparation
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Input: Wikipedia
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Output: a set of paragraphs
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
1. Obtaining and parsing of wikipedia dump
1. Selecting feasible paragraphs
Notes:
- PageRank Causes bias to geography, random selection might be the best
- [75 best articles](https://sk.wikipedia.org/wiki/Wikip%C3%A9dia:Zoznam_najlep%C5%A1%C3%ADch_%C4%8Dl%C3%A1nkov)
- [167 good articles](https://sk.wikipedia.org/wiki/Wikip%C3%A9dia:Zoznam_dobr%C3%BDch_%C4%8Dl%C3%A1nkov)
- [Wiki Facts](https://sk.wikipedia.org/wiki/Wikip%C3%A9dia:Zauj%C3%ADmavosti)
### Question Annotation
Input: A set of paragraphs
Output: A question for each paragraph
### Answer Annotation
Input: A set of paragraphs and questions
Output: An answer for each paragraph and question
### Annotation Summary
Annotation work summary
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Input: Database of annotations
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Output: Summary of work performed by each annotator
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
### Annotation Manual
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Output: Recommendations for annotators
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
### Question Answering Model
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Input: An annotated QA database
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Otput: An evaluated model for QA
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
Traing the model with annotated data:
2020-03-06 09:53:42 +00:00
2020-06-11 10:07:46 +00:00
- Selecting existing modelling approach
- Evaluation set selection
- Model evaluation
- Supporting the annotation with the model (pre-selecting answers)
### Supporting activities
Output: More annotations
Organizing voluntary student challenges to support the annotation process
## Existing implementations
- https://github.com/facebookresearch/DrQA
- https://github.com/brmson/yodaqa
- https://github.com/5hirish/adam_qas
- https://github.com/WDAqua/Qanary - metodológia a implementácia QA
## Bibligraphy
- Reading Wikipedia to Answer Open-Domain Questions, Danqi Chen, Adam Fisch, Jason Weston, Antoine Bordes
Facebook Research
- SQuAD: 100,000+ Questions for Machine Comprehension of Text https://arxiv.org/abs/1606.05250
## Existing Datasets
- Squad TheStanfordQuestionAnsweringDataset(SQuAD) (Rajpurkar et al., 2016)
- WebQuestions
- https://en.wikipedia.org/wiki/Freebase