知识图谱--笔记

启动:neo4j.bat console

CQL 增： 增加一个节点

create(n:Person{name:‘我’,age:31})

n:创建的节点名称

Person:节点标签名称

name:节点的属性的名称

value:节点的属性的值

带有关系属性

create(p:Person{name:‘我’,age:“31”})-[:包公程{金额：10000}]->(n:Person{name:“好大哥”,age:“35”})

MATCH (e:Customer),(cc:CreditCard) 
CREATE (e)-[r:DO_SHOPPING_WITH ]->(cc) 

创建带有多个标签的节点

CREATE (m:Movie:Cinema:Film:Picture)

删

create(n:Person{name:“TYD”,age:31})

match(n:Person{name:“TYD”})delete n

删除关系

match (p:Person{name:“我”,age:“31”})-[f:包工程]->(n:Person{name:“好大哥”,age:“35”})

delete f

MATCH (cc: CreditCard)-[rel]-(c:Customer) 
DELETE cc,c,rel

改

加上标签

match (t:Person) where id(t)=789 set t:好人return t

加上属性

match (a:好人) where id(a)=789 set a.战斗力=200 return a

修改属性

match (a:好人) where id(a)=789 set a.战斗力=500 return a

查

(查 *** 作太多啦，直接参考neo4j例子就好)

match (p:Person) - [:包工程] -> (n:Person) return p,n

MATCH (dept:Dept) return dept

MATCH (p:Employee {id:123,name:"Lokesh"}) RETURN p

MATCH (p:Employee) WHERE p.name = "Lokesh" RETURN p

MATCH (video1:YoutubeVideo1)-[movie:ACTION_MOVIES]->(video2:YoutubeVideo2) 
RETURN movie

布尔运算符：and,or,not,xor

比较运算符：=，<>(不等于)

MATCH (cust:Customer),(cc:CreditCard) 
WHERE cust.id = "1001" AND cc.id= "5001" 
CREATE (cust)-[r:DO_SHOPPING_WITH{shopdate:"12/12/2014",price:55000}]->(cc) 
RETURN r

快速清空数据库：

MATCH (n)

DETACH DELETE n

remove 永久删除节点的属性

MATCH (book { id:122 }) REMOVE book.price RETURN book

删除不需要的标签

MATCH (m:Movie)  REMOVE m:Picture

set子句

MATCH (book:Book) SET book.title = ‘superstar’ RETURN book

Sorting排序

order by（升序）

MATCH (emp:Employee)
RETURN emp.empid,emp.name,emp.salary,emp.deptno
ORDER BY emp.name

ORDER BY emp.name DESC（降序）

UNION联盟

MATCH (cc:CreditCard) RETURN cc.id,cc.number
UNION
MATCH (dc:DebitCard) RETURN dc.id,dc.number

因为xyk和借记卡有相同的属性名，但有不同的节点名称前缀

"as"子句用于给一些别名

MATCH (cc:CreditCard)
RETURN cc.id as id,cc.number as number,cc.name as name,
   cc.valid_from as valid_from,cc.valid_to as valid_to
UNION
MATCH (dc:DebitCard)
RETURN dc.id as id,dc.number as number,dc.name as name,
   dc.valid_from as valid_from,dc.valid_to as valid_to

在这里，因为UNION子句过滤它们，可以看到该命令返回9行没有重复的行

而使用union all命令返回10行

LIMIT和SKIP

limit来过滤或限制返回的行数（修剪底部）

skip（修剪顶部）

MATCH (emp:Employee) RETURN emp
LIMIT 2

MATCH (emp:Employee) RETURN emp
SKIP 2

merge合并

merge命令将新的节点添加到数据库，只有当它不存在

MERGE (gp2:GoogleProfile2{ Id: 201402,Name:"Nokia"})

NULL值

将空值视为对节点或关系属性的缺失值或未定义值

MATCH (e:Employee) 
WHERE e.id IS NULL
RETURN e.id,e.name,e.sal,e.deptno

IN *** 作符

为CQL命令提供值的集合

MATCH (e:Employee) 
WHERE e.id IN [123,124]
RETURN e.id,e.name,e.sal,e.deptno

ID属性

是节点和关系的默认内部属性

字符串函数 upper改为大写字母

MATCH (e:Employee) 
RETURN e.id,UPPER(e.name),e.sal,e.deptno

LOWER

MATCH (e:Employee) 
RETURN e.id,LOWER(e.name),e.sal,e.deptno

substring获取给定string的子字符串

substring(字符串，索引开始，索引结束)

MATCH (e:Employee) 
RETURN e.id,SUBSTRING(e.name,0,2),e.sal,e.deptno

aggregation聚合 count（行数）*，节点或关系标签名称或属性名称

MATCH (e:Employee) RETURN COUNT(*)

max，min,sum,avg

MATCH (e:Employee) 
RETURN MAX(e.sal),MIN(e.sal),SUM(e.sal),AVG(e.sal)

关系函数

startnode(关系的开始节点)，endnote，id（关系的id),type

MATCH (a)-[movie:ACTION_MOVIES]->(b) 
RETURN STARTNODE(movie),ENDNODE(movie),ID(movie),TYPE(movie)

索引

CREATE INDEX ON : ()
DROP INDEX ON : ()

CREATE INDEX ON :Customer (name)
DROP INDEX ON :Customer (name)

unique约束

（避免重复记录）

CREATE CONSTRAINT ON (cc:CreditCard)
ASSERT cc.number IS UNIQUE
在CreditCard节点的number属性上创建了一个unique约束

DROP CONSTRAINT ON ()
ASSERT  IS UNIQUE

Py2neo

link = Graph("http://localhost:7474", username="neo4j", password="tianzaichen1977")

1.创建节点与关系

from py2neo import Node,Relationship
a=Node("Person",name="Alice")
b=Node("Person",name="Bob")
ab=Relationship(a,"Knows",b)
//对Node或Relationship进行属性赋值
a['age']=20
b['age']=21
r['time']='2017/08/31'
//也可以通过setdefault()方法赋值默认属性
a.setdefault('location','北京')
//也可以用update()方法对属性批量更新
data={
'name':'Amy',
'age':21
}
a.update(data)

• hash(node) 返回node的ID的哈希值
• node[key] 返回node的属性值，没有此属性就返回None
• node[key] = value 设定node的属性值
• del node[key] 删除属性值，如果不存在此属性报KeyError
• len(node) 返回node属性的数量
• dict(node) 返回node所有的属性
• walk(node) 返回一个生成器且只包含一个node
• labels() 返回node的标签的集合
• has_label(label) node是否有这个标签
• add_label(label) 给node添加标签
• remove_label(label) 删除node的标签
• clear_labels() 清楚node的所有标签
• update_labels(labels) 添加多个标签，注labels为可迭代的

其中连接的属性有：

• hash(relationship) 返回一个关系的hash值
• relationship[key] 返回关系的属性值
• relationship[key] = value 设定关系的属性值
• del relationship[key] 删除关系的属性值
• len(relationship) 返回关系的属性值数目
• dict(relationship) 以字典的形式返回关系的所有属性
• walk(relationship) 返回一个生成器包含起始node、关系本身、终止node
• type() 返回关系type

1.2子图Subgraphs

s = a | b | r
print(s)
print(s.nodes())
print(s.relationships())
from py2neo import Node, Relationship, size, order
s = a | b | r
print(s.keys())
print(s.labels())
print(s.nodes())
print(s.relationships())
print(s.types())
print(order(s))
print(size(s))

>>> frozenset({'name'})
>>> frozenset({'Person'})
>>> frozenset({(alice:Person {name:"Alice"}), (bob:Person >>> >>> >>> {name:"Bob"})})
>>> frozenset({(alice)-[:KNOWS]->(bob)})
>>> frozenset({'KNOWS'})
>>> 2
>>> 1

• 并”|“

• 交”&“

• 差”-“

• 对称差”^"

• keys()返回子图节点和关系所有属性的集合

• labels()返回节点label的集合

• nodes()返回所有节点集合

• relationships()

• types()返回所有关系的type集合

• order(subgraph)返回子图节点的数目

• size(subgraph)返回子图关系的数目

1.3Walkable Types遍历功能的子图

walk()转换为一个生成器包含节点和关系

//将子图合并起来
from py2neo import Node, Relationship
 
a = Node('Person', name='Alice')
b = Node('Person', name='Bob')
c = Node('Person', name='Mike')
ab = Relationship(a, "KNOWS", b)
ac = Relationship(a, "KNOWS", c)
w = ab + Relationship(b, "LIKES", c) + ac
print(w)
>>> (alice)-[:KNOWS]->(bob)-[:LIKES]->(mike)<-[:KNOWS]-(alice)
//调用walk()方法实现遍历
from py2neo import walk
for item in walk(w):
    print(item)
'''
(alice:Person {name:"Alice"})
(alice)-[:KNOWS]->(bob)
(bob:Person {name:"Bob"})
(bob)-[:LIKES]->(mike)
(mike:Person {name:"Mike"})
(alice)-[:KNOWS]->(mike)
(alice:Person {name:"Alice"})
'''
//strat_node(),end_node(),nodes(),relationship()
print(w.start_node())
print(w.end_node())
print(w.nodes())
print(w.relationships())
'''
>>> (alice:Person {name:"Alice"})
>>> (alice:Person {name:"Alice"})
>>> ((alice:Person {name:"Alice"}), (bob:Person {name:"Bob"}), (mike:Person {name:"Mike"}), (alice:Person {name:"Alice"}))
>>> ((alice)-[:KNOWS]->(bob), (bob)-[:LIKES]->(mike), (alice)-[:KNOWS]->(mike))
'''

1.4链接已有图数据库.Graph()

//在database模块中包含了和Neo4j数据交互的API，最重要的是Graph
test_graph = Graph(
    "http://localhost:7474", 
    username="neo4j", 
    password="xxxx"
)
//利用create()方法传入subgraph对象将关系图，节点，关系添加到数据库中
graph=Graph("http://localhost:7474",username="neo4j",password="x")
graph.create(s)

1.5查询Query

graph.nodes.match()
graph = Graph()
matcher = NodeMatcher(graph)
matcher.match("Person", name="Keanu Reeves").first()
graph.match(nodes=None, r_type=None, limit=None) 
//用CQL进行查询，返回list
data1 = test_graph.run('MATCH (a:人) RETURN a')
print(data1, type(data1))
import pandas as pd
print(pd.DataFrame(graph.nodes.match('人')))
print(pd.DataFrame(graph.nodes.match('人', name='小刚')))
print(list(graph.match(r_type="喜欢")))

1.6删

graph.delete(a)
graph.delete_all()#全删

1.7改

a['age'] = 20
graph.push(a)