Knowledge Graph Querying

1. RDF Query Language: SPARQL

W3C Stack
SPARQL is an RDF query language, that is, a semantic query language for databases, able to retrieve and manipulate data stored in RDF format.

SPARQL stands for
- (originally) Simple Protocol and RDF Query Language.
- (now) SPARQL Protocol and RDF Query Language.
How to get information from RDF graphs by SPARQL?
- Pattern matching
  - Pattern: describe subgraphs of the queried RDF graph
  - Matching: match the pattern to the subgraphs to contribute an answer
  - Building blocks: graph patterns (i.e. RDF graphs with variables)

子图与知识图谱做匹配，把知识图谱中所有位于中心的词拿过来

1.1 SPARQL Example

1.2 Components of SPARQL Queries

Prologue
- Prefix definitions for compact URIs;
- Attention (difference w.r.t. Turtle): No period (“.”) character as separator.
Query Form
- SELECT, DESCRIBE, CONSTRUCT or ASK
Dataset specification
- From
- Specify the RDF dataset to be querying
Query pattern
- WHERE clause specifies the graph pattern to be matched
- 指明什么样的语句需要被匹配
Solution modifiers
- Order: put the solutions in order;
- Projection: choose certain variables;
  - 指定返回特定变脸结果
- Distinct: ensure solutions in the sequence are unique;
  - 确保返回结果是唯一
- Reduced: permit elimination of some non-unique solutions;
  - 防止删除重复
- Offset: control where the solutions start from in the overall sequence of solutions;
  - 从第n个返回
- Limit: restrict the number of solutions.
  - 限制返回数量

2. SPARQL

SPARQL Syntax (RDF term syntax)
- Syntax for IRI
- Syntax for literals
- Syntax for variables
- Syntax for blank nodes
- Graph Patterns for Query Pattern
- Triple Pattern
- Different Graph Patterns

2.1 SPARQL Syntax: RDF Term Syntax

Syntax for IRI
- IRIs are a generalization of URIs and are fully compatible with URIs and URLs.
- The following fragments are some of the different ways to write the same IRI:

一旦定义BASE，那么所有的都是这一个；PREFIX则不同
The general syntax for literals
- A string (enclosed in either double quotes, “…”, or single quotes, ‘…’);
- With either an optional language tag (introduced by @) or an optional datatype IRI or prefixed name (introduced by ^^).

字符串用单引号或双引号皆可
Syntax for query variables
- Query variables in SPARQL queries have the global scope
  - Use of a given variable name anywhere in a query identifies the same variable.
- Variables are prefixed by either “?” or “\$”;
  - The “?” or “\$” is not part of the variable name.

查找所有谓语是name的三元组，返回其并于
Syntax for blank nodes
- 空节点就是表示变量
- Blank nodes in graph patterns act as non-distinguished variables, not as references to specific blank nodes in the data being queried.
- Blank nodes are indicated by either the label form, such as “_:abc”, or the abbreviated form “[]”.
  - A blank node that is used in only one place in the query syntax can be indicated with [].
- The same blank node label cannot be used in two different basic graph patterns in the same query.

Triple Patterns
- are basic units of graph patterns;
- are written as a whitespace-separated list of a subject, predicate and object;
- There are abbreviated ways of writing some common triple patterns;
- The following examples express the same query:

Triple Patterns: Predicate-Object Lists

Triple patterns with a common subject can be written so that the subject is only written once and is used for more than one triple pattern by employing the “;” notation.
共用主语

Exercise

Please rewrite the following triple pattern by using the same subject only once.

1
2
3

?a int:number 123789 .
?a int:pair 34567.
?a int:id 666777 .

1
2
3

?a int:number 123789；
   int:pair 34567；
   int:id 666777 .

Triple Patterns: Object Lists
- If triple patterns share both subject and predicate, the objects may be separated by “,”.

Exercise

Please rewrite the following triple pattern by using the same subject and predicate only once.

?a string:name ‘Bob’.
?a string:name ’Bobby’.
?a string:name ‘Boob’.
?a string:name ‘Bob_’.

1	?a string:name ‘Bob’,’Bobby’,‘Boob’,‘Bob_’.

2.2 Graph Patterns

SPARQL is based around graph pattern matching.
Different types of graph patterns for the query pattern (WHERE clause):
- Basic Graph Patterns, where a set of triple patterns must match;
- Group Graph Pattern, where a set of graph patterns must all match;
- Optional Graph patterns, where additional patterns may extend the solution;
- Alternative (Union) Graph Pattern, where two or more possible patterns are tried;

2.2.1 Basic graph patterns

Triple patterns are similar to RDF triples, but any component can be a
query variable.

1	?x foaf:name ?name .

Matching a triple pattern to a graph: bindings between variables and
RDF terms.
Matching of basic graph patterns:
- A Pattern Solution of Graph Pattern GP on graph G is any substitution S such that S(GP) is a subgraph of G.
  - Simple queries
  - Multiple matches
  - Matching RDF literals
  - Blank node labels in query results
Simple queries

注意返回要有列名

Exercise
Given the dataset and the SPARQL query as follows, please write the query results.

# Graph: http://example/addresses
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
<http://example/president25> foaf:givenName "Bill" .
<http://example/president25> foaf:familyName "McKinley" .
<http://example/president27> foaf:givenName "Bill" .
<http://example/president27> foaf:familyName "Taft" .
<http://example/president42> foaf:givenName "Bill" .
<http://example/president42> foaf:familyName "Clinton" .

PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?name ?family
WHERE { ?x foaf:givenName ?name .
?x foaf:familyName ?family .}

name	family
“Bill”	“McKinley”
“Bill”	“Taft”
“Bill”	“Clinton”

Multiple Matches

Exercise

Given the dataset and the SPARQL query as follows, please write the query results.

# Graph: http://example/addresses
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
<http://example/president25> foaf:givenName "Bill" .
<http://example/president25> foaf:familyName "McKinley" .
<http://example/president27> foaf:givenName “Bob" .
<http://example/president27> foaf:id 159486.
<http://example/president42> foaf:givenName “Marry" .
<http://example/president42> foaf:familyName "Clinton" .

PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?name ?family
WHERE { ?x foaf:givenName ?name .
?x foaf:familyName ?family .}

name	family
“Bill”	“McKinley”
“Marry”	“Clinton”

Matching RDF literals

@prefix dt: <http://example.org/datatype#> .
@prefix ns: <http://example.org/ns#> .
@prefix : <http://example.org/ns#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
:x ns:p "cat"@en .
:y ns:p "42"^^xsd:integer .
:z ns:p "abc"^^dt:specialDatatype .

第一种是找不到结果的，因为不加@是不一样的意义

Blank node labels in query results
- There need not be any relation between a label in the form of a blank node in the result set and a blank node in the data graph with the same label.
Blank node labels in query results
- There need not be any relation between a label in the form of a blank node in the result set and a blank node in the data graph with the same label.

2.2.2 Group Graph Patterns

In a SPARQL query string, a group graph pattern is delimited with braces: {}.

Besides triple patterns, group graph pattern can contain constraints
- Syntax: Keyword FILTER followed by a filter expression

name	family
“Bill”	“McKinley”

2.2.3 Optional Graph Patterns

If the optional part does not match, it creates no bindings but does not eliminate the solution.
Optional patterns may result in unbound variables

Exercise
- Given the dataset and the SPARQL query as follows, please write the query results.

Answer

name	family
“Bill”	“McKinley”
“Bob”
“Marry”

Constraints in Optional Pattern Matching

Multiple Optional Graph Patterns

2.2.4 Alternative(Union) Graph Patterns

Combine graph patterns so that one of several alternative graph patterns may match.

Exercise

Please write the SPARQL query on “list all volcanos located in Italy or Norway” given the following data.

depedia:Mount_Etna	rdf:type	umbel-sc:Volcano;
					rdfs:label 	"Etna";
					p:location	dbpedia:Italy.
depedia:Mount_Baker	rdf:type	umbel-sc:Volcano;
					p:location	dbpedia:United_States.
depedia:Beerenberg	rdf:type	umbel-sc:Volcano;
					rdfs:label	"Beerenberg"@en;
					p:location	dbpedia:Norway.

Answer

SELECT ?volcano rdf:type umbel-sc:Volcano.
WHERE {
    {?Mount p:location dbpedia:Italy.}
    UNION {?Mount p:location dbpedia:Norway.}
}

2.3 Dataset specification

SPARQL queries are executed over an RDF dataset:
- One default graph and
- Zero or more named graphs (identified by an IRI).
Evaluation of patterns w.r.t. the active graph (initially the default graph),
i.e., the graph used for matching graph patterns;
GRAPH clause is used for making a named graph the active graph.

加了Graph 子句就会在那么graph查询，否则就会在defalt查询

2.4 Query Forms

SELECT
- Result: sequence of solutions (i.e., sets of variable bindings);
- Selected variables separated by space (not by comma!);
- Asterisk character (“*”) selects all variables in the pattern.

Exercise

Given the dataset and the SPARQL query as follows, please write the query results.
Dataset

# Default graph (stored at http://example.org/dft.ttl)
@prefix dc: <http://purl.org/dc/elements/1.1/> .
<http://example.org/bob> dc:publisher "Bob Hacker" .
<http://example.org/alice> dc:publisher "Alice Hacker" .

# Named graph: http://example.org/bob
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
_:a foaf:name "Bob" .
_:a foaf:mbox <mailto:bob@oldcorp.example.org> .

# Named graph: http://example.org/alice
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
_:a foaf:name "Alice" .
_:a foaf:mbox <mailto:alice@work.example.org> .

SPARQL query
- FROMNAMD可以省略，因为标明了GRAPH ?g

PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX dc: <http://purl.org/dc/elements/1.1/>
SELECT ?who ?g ?mbox
FROM <http://example.org/dft.ttl>
FROM NAMED <http://example.org/alice>
FROM NAMED <http://example.org/bob>
WHERE
{
?g dc:publisher ?who .
GRAPH ?g { ?x foaf:mbox ?mbox }
}

Answer:
| who | g | mbox |
| ——— | ————— | ————— |
| “Bob Hacker” | \http://example.org/bob | \bob@oldcorp.example.org\|
| “Alice Hacker” | \http://example.org/alice | |

2.5 DESCRIBE

Result: an RDF graph (i.e., all RDF triples) that describes the resources found;
The DESCRIBE clause can take IRIs to identify the resources.
The resources to be described can also be taken from the bindings to a query variable in a result set.

2.6 CONSTRUCT

Result: an RDF graph constructed from a template;
Template: a graph pattern with the variables from the query pattern.
将模板的变量换掉，其余不变

dbpedia:Mount_Etna rdfs:label “Etna”;

rdf:type myTypes:VolcanosOutsideTheUS.
dbpedia:Beerenberg rdfs:label “Beerenberg”@en;

rdf:type myTypes:VolcanosOutsideTheUS.

2.7 ASK

Check whether there is at least one result;
Result: true or false.

2.8 DELETE/INSERT

INSERT
- Insert the new RDF triples into the existing RDF graph.

DELETE
- Delete some triples in the RDF graph.

DELETE/INSERT
- Remove or add triples from/to the Graph Store based on bindings for a query pattern specified in a where clause.
WITH 表示数据集
先做查询，在再查询里做插入删除

Exercise

Given the dataset and the SPARQL query as follows, please write the query results.

Q1

@prefix org: <http://example.com/ns#> .
_:a org:employeeName "Alice" .
_:a org:employeeId 12345 .
_:b org:employeeName "Bob" .
_:b org:employeeId 67890 .

PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX org: <http://example.com/ns#>
DELETE {?person ?p ?o .}
where {?person org:employeeId ?id .
FILTER (?id > 50000)
?person ?p ?o .}

删除后：

1 2	_:a org:employeeName "Alice" . _:a org:employeeId 12345 .

2. Q2

@prefix org: <http://example.com/ns#> .
# Graph: http://person
_:a org:employeeName "Alice" .
_:a org:employeeId 12345 .
_:b org:employeeName "Bob" .
_:b org:employeeId 67890 .
# Graph: http://person2
_:c org:employeeId 13579

PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX org: <http://example.com/ns#>
INSERT {GRAPH <http://person2> {?person ?p ?o .}}
where {Graph: <http://person>
{?person org:employeeId ?id .
FILTER (?id < 50000)
?person ?p ?o .}}

@prefix org: <http://example.com/ns#> .
# Graph: http://person
_:a org:employeeName "Alice" .
_:a org:employeeId 12345 .
_:b org:employeeName "Bob" .
_:b org:employeeId 67890 .
# Graph: http://person2
_:c org:employeeId 13579 .
_:a org:employeeName "Alice" .
_:a org:employeeId 12345 .

2.9 CLEAR

Remove all the triples in the specified graph(s) in the Graph Store.

2.10 MOVE

Move all data from an input graph into a destination graph.

2.11 Solution Modifiers

Only for SELECT queries;
Modify the result set as a whole (not single solutions);
Keywords: DISTINCT, ORDER BY, LIMIT, OFFSET.

DISTINCT

Remove duplicates from the result set.

ORDER BY

Order the results.

ASC for ascending (default) and DESC (e.g., DESC(?name)) for descending.

LIMIT

limits the number of result:
- 只返回5个

OFFSET

position/index of the first reported results:

Order of the result should be predictable (combine with ORDER BY)

BINDINGS

VALUE

add data to the query directly.
- 增加限制

AGGREGATES

allows for the grouping of solutions and
the computation of values over the groups.

GROUP BY groups the solutions; (i.e., students who attend the same lecture)
COUNT is an aggregate function that counts the solutions within a group; (i.e., number of students in the lecture)
HAVING filters aggregated values
Question: Please use natural language to explain this SPARQL query!
- 查选课人数超过5人的课

NEGATION:

Question: Please use natural language to explain the above SPARQL queries!