Merge branch 'master' of git.kemt.fei.tuke.sk:KEMT/zpwiki

This commit is contained in:
Daniel Hládek 2022-10-06 14:02:56 +02:00
commit a0eebc2829
2 changed files with 63 additions and 0 deletions

View File

@ -0,0 +1,4 @@
# Dense Passage Retrieval for Open-Domain Question Answering
## Clanok č.1
### Dense Passage Retriever (DPR)

View File

@ -37,6 +37,39 @@ Goals:
- Publish 1 regular conference paper. - Publish 1 regular conference paper.
- Prepare a demo for hate speech detection. - Prepare a demo for hate speech detection.
Meeting 5.10.
Status:
Studied python and ML:
- Basics to Advanced python is completed
- word2vec and word embedding examples are tried
- Basic tools (tensorflow) on Machine Learning and task are completed
- learning on ML and Deep learning libraries and fairseq RNN, SVM BERT code samples
Building the datasets memes based on Dravidian languages:
- Collected base papers on Hate speech Multilanguage troll and not troll memes with low resource languages
Currected the survey paper and shared for review.
Worked on the baseline HS expriment.
- https://git.kemt.fei.tuke.sk/mg240ia/Hate-Speech-Detector-Streamlit
Tasks:
- Publish a paper on SAMI 2023
- publish a paper on a school conference, ask for deadline.
Possible paper topics:
- contine to work on the baseline HS experiment. Evaluate accuracy for the classifiers. This is a possible simple publication.
- continue to work on the horsehead experiment. This is another possible paper.
- continue work on the dravidian dataset. this in another possible papers.
- continue to work on the survey paper. This is another possible Q3 paper.
Meeting 6.9.2022 Meeting 6.9.2022
Status: Status:
@ -72,6 +105,32 @@ Resources:
- [Resources and benchmark corpora for hate speech detection: a systematic review] - [Resources and benchmark corpora for hate speech detection: a systematic review]
(https://link.springer.com/article/10.1007/s10579-020-09502-8) (https://link.springer.com/article/10.1007/s10579-020-09502-8)
Meeting 5.10.
Status:
Studied python and ML:
- Basics to Advanced python is completed
- word2vec and word embedding examples are tried
- Basic tools (tensorflow) on Machine Learning and task are completed
- learning on ML and Deep learning libraries and fairseq RNN, SVM BERT code samples
Building the datasets memes based on Dravidian languages:
- Collected base papers on Hate speech Multilanguage troll and not troll memes with low resource languages
Currected the survey paper and shared for review.
Worked on a baseline HS experiment:
https://git.kemt.fei.tuke.sk/mg240ia/Hate-Speech-Detector-Streamlit
Tasks:
- Publish a paper about the dataset on the SAMI 2023 Conference.
- Publish a paper in school journal - ask about deadlilne.
14.7: 14.7:
Status: Status: