|
Semantic queries allow for queries and analytics of associative and contextual nature. Semantic queries enable the retrieval of both explicitly and implicitly derived information based on syntactic, semantic and structural information contained in data. They are designed to deliver precise results (possibly the distinctive selection of one single piece of information) or to answer more fuzzy and wide open questions through pattern matching and digital reasoning. Semantic queries work on named graphs, linked-data or triples. This enables the query to process the actual relationships between information and ''infer'' the answers from the ''network of data''. This is in contrast to semantic search, which uses semantics (the science of meaning) in unstructured text to produce a better search result (see Natural language processing). From a technical point of view semantic queries are precise relational-type operations much like a database query. They work on structured data and therefore have the possibility to utilize comprehensive features like operators (e.g. >, < and =), namespaces, pattern matching, subclassing, transitive relations, semantic rules and contextual full text search. The semantic web technology stack of the W3C is offering SPARQL〔(【引用サイトリンク】title=Introducing SPARQL: Querying the Semantic Web )〕〔(【引用サイトリンク】title=SPARQL Query Language for RDF )〕 to formulate semantic queries in a syntax similar to SQL. Semantic queries are used in triplestores, graph databases, semantic wikis, natural language and artificial intelligence systems. == Background == Relational databases contain all relationships between data in an ''implicit'' manner only.〔(【引用サイトリンク】title=Semantic queries in databases: problems and challenges )〕〔(【引用サイトリンク】title=Karma: A System for Mapping Structured Sources into the Semantic Web )〕 For example the relationships between customers and products (stored in two content-tables and connected with an additional link-table) only come into existence in a query statement (SQL in the case of relational databases) written by a developer. Writing the query demands the exact knowledge of the database schema.〔(【引用サイトリンク】title=A Scalable Approach to Learn Semantic Models of Structured Sources )〕〔(【引用サイトリンク】title=Semantics for Big Data Integration and Analysis )〕 Linked-Data contain all relationships between data in an ''explicit'' manner. In the above example no query code needs to be written. The correct product for each customer can be fetched automatically. Whereas this simple example is trivial, the real power of linked-data comes into play when a ''network of information'' is created (customers with their geo-spatial information like city, state and country; products with their categories within sub- and super-categories). Now the system can automatically answer more complex queries and analytics that look for the connection of a particular location with a product category. The development effort for this query is omitted. Executing a semantic query is conducted by ''walking'' the network of information and finding matches (also called ''Data Graph Traversal''). Another important aspect of semantic queries is that the type of the relationship can be used to incorporate intelligence into the system. The relationship between a customer and a product has a fundamentally different nature then the relationship between a neighbourhood and its city. The latter enables the semantic query engine to ''infer'' that a customer ''living in Manhattan is also living in New York City'' whereas other relationships might have more complicated patterns and "contextual analytics". This process is called inference or reasoning and is the ability of the software to derive new information based on given facts. 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「semantic query」の詳細全文を読む スポンサード リンク
|