在neo4j画图之旅

1.启动在12服务器上的neo4j

neo4j status
neo4j start

启动服务器上的neo4j 启动成功

2.数据集样例(知识库中的三元组)

数据形式：头实体\t关系\t尾实体 ,并将该数据放在train.txt文件中

Andrei_Tarkovsky    birthPlace  Soviet_Union
Ayn_Rand    citizenship Soviet_Union
Albert_Einstein deathPlace  Princeton,_New_Jersey
Albert_Einstein award   Nobel_Prize_in_Physics
Amsterdam   part    Amsterdam-Zuid
ABBA    recordLabel RCA_Records
ABBA    associatedBand  Benny_Anderssons_orkester
ABBA    associatedMusicalArtist Benny_Anderssons_orkester
Aarhus  governmentType  Magistrate
Aarhus  isPartOf    Central_Denmark_Region
Albert,_Duke_of_Prussia birthPlace  Ansbach
Abadan,_Iran    country Iran

3.三元组写入neo4j代码

# -*- encoding: utf-8 -*-
import os
from py2neo import *

class EstablishConceptGraph(object):
    """
    Build graph
    Node and Relation
    """
    def __init__(self,data_file='train.txt',data_name='dbpedia50'):
        self.graph = Graph(host='localhost', auth=('neo4j', '123'))
        self.dataf = data_file
        self.dataname = data_name
        self.conceptNodeDict = self.freading()

    def freading(self):
        nodeSet = set()
        triples = []
        with open(self.dataf,'r',encoding='utf8') as f:
            line = f.readline().strip()
            while line:
                t = line.split('\t')
                nodeSet.add(t[0])
                nodeSet.add(t[2])
                triples.append(t)
                line=f.readline().strip()

        entity_node = list(nodeSet)
        entity_node.sort()

        conceptNodeDict = dict()
        for i in range(len(entity_node)):
            # print('concept'+str(i+1),entity_node[i])
            idStr = 'concept'+str(i+1)
            conceptNodeDict[entity_node[i]] = idStr

        for t in triples:
            # print(t[0], t[1], t[2])
            # print(conceptNodeDict[t[0]],conceptNodeDict[t[2]])
            n1 = Node(self.dataname, name=t[0], id=conceptNodeDict[t[0]])
            n2 = Node(self.dataname, name=t[2], id=conceptNodeDict[t[2]])
            tx = self.graph.begin()
            rel = Relationship(n1,t[1],n2)
            tx.merge(n1, self.dataname, "id")  # node,label,primary key
            tx.merge(n2, self.dataname, "id")  # node,label,pirmary key
            tx.merge(rel)
            tx.commit()
        return conceptNodeDict

if __name__ == "__main__":
    conceptGraph = EstablishConceptGraph()

写入成功！

4.清除所有内容

MATCH (cc: dbpedia50)-[r]-(e:dbpedia50) 
DELETE cc,e,r

刚才添加的数据清除成功