Disclaimer: This article is a blogger original article, follow the CC 4.0 BY-SA copyright agreement, reproduced, please attach the original source link and this statement.
This link: https: //blog.csdn.net/m0_38106923/article/details/102805399
recently, has been working knowledge of OwnThink platform mapping studies on the history of the largest open source Github 140 million Chinese knowledge map, where the data is ( entity, attribute, value), (entities, relationships, entity) a mixed form of organization, the data format employed csv format.
So far, OwnThink project open dialogue robot, knowledge maps, semantic understanding, natural language processing tools. Knowledge Mapping integration of over 2,500 million for the entity, the entity owns property relations one hundred million level, the robot uses a semantic perception and understanding based on mapping of knowledge, committed to the strongest cognitive brain. Natural Language Processing Toolkit features include: Chinese word segmentation, POS tagging, named entity recognition, keyword extraction, text summarization, discover new words, sentiment analysis and so on.
Rolling a variety of keywords on the home page OwnThink platform, users can input their knowledge looking for, then you can draw the appropriate knowledge map.
OwnThink knowledge map can also be applied to a robot answering system, knowledge is recommended and so on. Application of knowledge on the robot map below shows.
OwnThink support online API calls, little interested partners can use for measurement, using the Python dialogue robot simple call test.
import json
import requests
sess = requests.get('https://api.ownthink.com/bot?spoken=中国')
answer = sess.text
answer = json.loads(answer)
print(answer)
效果如下：
GitHub address: https: //github.com/ownthink/KnowledgeGraphData
OwnThink website address: https: //www.ownthink.com/
----------------
Disclaimer: This article is CSDN bloggers "No hair loss program ape" of the original article, follow the CC 4.0 BY-SA copyright agreement, reproduced, please attach the original source link and this statement.
Original link: https: //blog.csdn.net/m0_38106923/article/details/102805399