關(guān)鍵詞作為一個對文本常用的概括,可以被應(yīng)用于關(guān)鍵詞云計算等應(yīng)用上。BosonNLP 的關(guān)鍵詞提取引擎可以將文本自動進(jìn)行關(guān)鍵詞分析,給出每個詞語相應(yīng)的權(quán)重。
關(guān)鍵詞提取引擎的調(diào)用很簡單,只需要將文本上傳到分析服務(wù)器,引擎會自動返回每個詞語的權(quán)重。
Note
注意這些權(quán)重的平方和為 1。
$ curl -X POST -H "Content-Type: application/json" -H "Accept: application/json" -H "X-Token: YOUR_API_TOKEN" --data "\"\u75c5\u6bd2\u5f0f\u5a92\u4f53\u7f51\u7ad9\uff1a\u8ba9\u65b0\u95fb\u8fc5\u901f\u8513\u5ef6\"" http://api.bosonnlp.com/keywords/analysis?top_k=2[[0.4580507649282757, "\u8513\u5ef6"], [0.44467176143180404, "\u75c5\u6bd2"]]
# -*- encoding: utf-8 -*-from __future__ import print_function, unicode_literalsimport jsonimport requestsKEYWORDS_URL = 'http://api.bosonnlp.com/keywords/analysis'text = '病毒式媒體網(wǎng)站:讓新聞迅速蔓延'params = {'top_k': 10}data = json.dumps(text)headers = { 'X-Token': 'YOUR_API_TOKEN', 'Content-Type': 'application/json'}resp = requests.post(KEYWORDS_URL, headers=headers, params=params, data=data.encode('utf-8'))for weight, word in resp.json(): print(weight, word)
$ python keywords_api_example.py0.458050764928 蔓延0.444671761432 病毒0.377757036473 迅速0.345847781395 網(wǎng)站0.341803722579 媒體0.315811136856 式0.305528321034 新聞0.142237269898 讓
# -*- encoding: utf-8 -*-from __future__ import print_function, unicode_literalsfrom bosonnlp import BosonNLP# 注意:在測試時請更換為您的 API tokennlp = BosonNLP('YOUR_API_TOKEN')s = '病毒式媒體網(wǎng)站:讓新聞迅速蔓延'result = nlp.extract_keywords(s, top_k=10)for weight, word in result: print(weight, word)
詳細(xì)的 Python SDK 關(guān)鍵詞提取文檔請看 這里 。