ccc
dedb5e01c5
add README.md
|
пре 3 година | |
|---|---|---|
| .ipynb_checkpoints | пре 3 година | |
| stoplist | пре 3 година | |
| README.md | пре 3 година | |
| cn_stopwords.txt | пре 3 година | |
| customized_stopwords.pickle | пре 3 година | |
| customized_stopwords.txt | пре 3 година | |
| dict.txt.big | пре 3 година | |
| gnews_keyword_extraction.ipynb | пре 3 година | |
| gnews_keyword_extraction.py | пре 3 година | |
| jieba_add_word.txt | пре 3 година | |
| jieba_add_word_kw_with_weighting.txt | пре 3 година | |
| predict_doc.txt | пре 3 година | |
| renewhouse_list.pickle | пре 3 година | |
| requirements.txt | пре 3 година | |
| tag_list.csv | пре 3 година |
README.md
Gnews Keyword Extraction
Extract keywords in target domain news from DB (gnews.gnews_detail).
First, we use Transformer to get the news semantic vector, and use HDBSCAN for clustering. Then we can use predict_doc.txt to predict which cluster the target domain may be in. Finally, we used unsupervised algorithms such as RAKE, TF-IDF, TextRank, and MultipartiteRank to extract the keywords of news in this cluster and store them in tag_list.csv.
Install
pip install -r requirements.txt
Note: See requirements.txt for more details.
Usage
Step 1
Copy the news content of the target domain and paste it to predict_doc.txt.
Step 2
python gnews_keyword_extraction.py --topK 80 --target_domain_range 1
--topK(optional) : – Get the top K keywords. (Default:80)--target_domain_range(optional) : – Candidate range of target domain.0means only one target cluster is used,1means a total of three clusters are used from the side. (Default:1)
Step 3
Get keywords from tag_list.csv.