Complex Data Types PDF
Document Details
Uploaded by Deleted User
null
null
Silberschatz, Korth and Sudarshan
Tags
Summary
This document is a chapter from a textbook on database system concepts. It provides a detailed overview of complex data types, including semi-structured data, object orientation, textual and spatial data. It covers JSON, XML, and RDF representation, as well as how to query such data.
Full Transcript
Chapter 8: Complex Data Types Database System Concepts, 7th Ed. ©Silberschatz, Korth and Sudarshan www.db-book.com See www.db-book.com for conditions on re-use Outline ▪ Semi-Structured Data ▪ Object Or...
Chapter 8: Complex Data Types Database System Concepts, 7th Ed. ©Silberschatz, Korth and Sudarshan www.db-book.com See www.db-book.com for conditions on re-use Outline ▪ Semi-Structured Data ▪ Object Orientation ▪ Textual Data ▪ Spatial Data Database System Concepts - 7th Edition 8.2 ©Silberschatz, Korth and Sudarshan Semi-Structured Data ▪ Many applications require storage of complex data, whose schema changes often ▪ The relational model’s requirement of atomic data types may be an overkill E.g., storing set of interests as a set-valued attribute of a user profile may be simpler than normalizing it ▪ Data exchange can benefit greatly from semi-structured data Exchange can be between applications, or between back-end and front-end of an application Web-services are widely used today, with complex data fetched to the front-end and displayed using a mobile app or JavaScript ▪ JSON and XML are widely used semi-structured data models Database System Concepts - 7th Edition 8.3 ©Silberschatz, Korth and Sudarshan Features of Semi-Structured Data Models ▪ Flexible schema Wide column representation: allow each tuple to have a different set of attributes, can add new attributes at any time Sparse column representation: schema has a fixed but large set of attributes, by each tuple may store only a subset ▪ Multivalued data types Sets, multisets ▪ E.g.,: set of interests {‘basketball, ‘La Liga’, ‘cooking’, ‘anime’, ‘jazz’} Key-value map (or just map for short) ▪ Store a set of key-value pairs ▪ E.g., {(brand, Apple), (ID, MacBook Air), (size, 13), (color, silver)} ▪ Operations on maps: put(key, value), get(key), delete(key) , Arrays ▪ Widely used for scientific and monitoring applications Database System Concepts - 7th Edition 8.4 ©Silberschatz, Korth and Sudarshan Features of Semi-Structured Data Models ▪ Arrays Widely used for scientific and monitoring applications E.g., readings taken at regular intervals can be represented as array of values instead of (time, value) pairs ▪ [5, 8, 9, 11] instead of {(1,5), (2, 8), (3, 9), (4, 11)} ▪ Multi-valued attribute types Modeled using non first-normal-form (NFNF) data model Supported by most database systems today ▪ Array database: a database that provides specialized support for arrays E.g., compressed storage, query language extensions etc Oracle GeoRaster, PostGIS, SciDB, etc Database System Concepts - 7th Edition 8.5 ©Silberschatz, Korth and Sudarshan Nested Data Types ▪ Hierarchical data is common in many applications ▪ JSON: JavaScript Object Notation Widely used today ▪ XML: Extensible Markup Language Earlier generation notation, still used extensively Database System Concepts - 7th Edition 8.6 ©Silberschatz, Korth and Sudarshan JSON ▪ Textual representation widely used for data exchange ▪ Example of JSON data { "ID": "22222", "name": { "firstname: "Albert", "lastname: "Einstein" }, "deptname": "Physics", "children": [ {"firstname": "Hans", "lastname": "Einstein" }, {"firstname": "Eduard", "lastname": "Einstein" } ] } ▪ Types: integer, real, string, and Objects: are key-value maps, i.e. sets of (attribute name, value) pairs Arrays are also key-value maps (from offset to value) Database System Concepts - 7th Edition 8.7 ©Silberschatz, Korth and Sudarshan JSON ▪ JSON is ubiquitous in data exchange today Widely used for web services Most modern applications are architected around on web services ▪ SQL extensions for JSON types for storing JSON data Extracting data from JSON objects using path expressions ▪ E.g. V-> ID, or v.ID Generating JSON from relational data ▪ E.g. json.build_object(‘ID’, 12345, ‘name’, ‘Einstein’) Creation of JSON collections using aggregation ▪ E.g. json_agg aggregate function in PostgreSQL Syntax varies greatly across databases ▪ JSON is verbose Compressed representations such as BSON (Binary JSON) used for efficient data storage Database System Concepts - 7th Edition 8.8 ©Silberschatz, Korth and Sudarshan XML ▪ XML uses tags to mark up text ▪ E.g. CS-101 Intro. to Computer Science Comp. Sci. 4 CS-102 Intro. to programming Science Comp. Sci. 4 Tags make the data self-documenting ▪ Tags can be hierarchical Database System Concepts - 7th Edition 8.9 ©Silberschatz, Korth and Sudarshan Example of Data in XML ▪ P-101 Cray Z. Coyote Route 66, Mesa Flats, Arizona 86047, USA Acme Supplies 1 Broadway, New York, NY, USA RS1 Atom powered rocket sled 2 199.95 … 429.85 …. Database System Concepts - 7th Edition 8.10 ©Silberschatz, Korth and Sudarshan XML Cont. ▪ XQuery language developed to query nested XML structures Not widely used currently ▪ SQL extensions to support XML Store XML data Generate XML data from relational data Extract data from XML data types ▪ Path expressions ▪ See Chapter 30 (online) for more information Database System Concepts - 7th Edition 8.11 ©Silberschatz, Korth and Sudarshan Knowledge Representation ▪ Representation of human knowledge is a long-standing goal of AI Various representations of facts and inference rules proposed over time ▪ RDF: Resource Description Format Simplified representation for facts, represented as triples (subject, predicate, object) ▪ E.g., (NBA-2019, winner, Raptors) (Washington-DC, capital-of, USA) (Washington-DC, population, 6,200,000) Models objects that have attributes, and relationships with other objects ▪ Like the ER model, but with a flexible schema ▪ (ID, attribute-name, value) ▪ (ID1, relationship-name, ID2) Has a natural graph representation Database System Concepts - 7th Edition 8.12 ©Silberschatz, Korth and Sudarshan Graph View of RDF Data ▪ Knowledge graph Database System Concepts - 7th Edition 8.13 ©Silberschatz, Korth and Sudarshan Triple View of RDF Data Database System Concepts - 7th Edition 8.14 ©Silberschatz, Korth and Sudarshan Querying RDF: SPARQL ▪ Triple patterns ?cid title "Intro. to Computer Science" ?cid title "Intro. to Computer Science" ?sid course ?cid ▪ SPARQL queries select ?name where { ?cid title "Intro. to Computer Science". ?sid course ?cid. Hint: course is a relationship (predicate) ?id takes ?sid. indicating that the student (?sid) is enrolled ?id name ?name. or associated with the course (?cid) } Also supports ▪ Aggregation, Optional joins (similar to outerjoins), Subqueries, etc. ▪ Transitive closure on paths Database System Concepts - 7th Edition 8.15 ©Silberschatz, Korth and Sudarshan Example: For the following example , write a SPARQL query to retrieve the titles of the books along with their authors' names? Solution : SELECT ?bookTitle ?authorName WHERE { ?book a :Book ; :title ?bookTitle ; :author ?author. ?author a :Person ; :name ?authorName. } Database System Concepts - 7th Edition 8.16 ©Silberschatz, Korth and Sudarshan RDF Representation (Cont.) ▪ RDF triples represent binary relationships ▪ How to represent n-ary relationships? Approach 1 (from Section 6.9.4): Create artificial entity, and link to each of the n entities ▪ E.g., (Barack Obama, president-of, USA, 2008-2016) can be represented as (e1, person, Barack Obama), (e1, country, USA), (e1, president-from, 2008) (e1, president-till, 2016) Approach 2: use quads instead of triples, with context entity ▪ E.g., (Barack Obama, president-of, USA, c1) (c1, president-from, 2008) (c1, president-till, 2016) ▪ RDF widely used as knowledge base representation DBPedia, Yago, Freebase, WikiData,.. ▪ Linked open data project aims to connect different knowledge graphs to allow queries to span databases Database System Concepts - 7th Edition 8.17 ©Silberschatz, Korth and Sudarshan Object Orientation ▪ Object-relational data model provides richer type system with complex data types and object orientation ▪ Applications are often written in object-oriented programming languages Type system does not match relational type system Switching between imperative language (Procedural) and SQL (Declarative) is troublesome. ▪ Approaches for integrating object-orientation with databases Build an object-relational database, adding object-oriented features to a relational database ( Ex.inherit Person table details to employee table). Automatically convert data between programming language model and relational model; data conversion specified by object-relational mapping (Mapping classes to tables). Build an object-oriented database that natively supports object- oriented data and direct access from programming language. Database System Concepts - 7th Edition 8.18 ©Silberschatz, Korth and Sudarshan Object-Relational Database Systems ▪ create user-defined types (UDTs) and use them to define a table structure : ▪ User-defined types - New UDT called Person create type Person (ID varchar(20) primary key, name varchar(20), address varchar(20)) ref from(ID); create table people of Person; the people table will inherit the structure ▪ Table types of the Person type, meaning it will have Columns ID, name, and address as defined create type interest as table ( the Person type. topic varchar(20), degree_of_interest int); create table users ( ID varchar(20), name varchar(20), interests interest); ▪ Array, multiset data types also supported by many databases Syntax varies by database Database System Concepts - 7th Edition 8.19 ©Silberschatz, Korth and Sudarshan Type and Table Inheritance ▪ Type inheritance create type Student under Person (degree varchar(20)) ; create type Teacher under Person (salary integer); ▪ Table inheritance syntax in PostgreSQL (inherits ) and oracle (UNDER) create table students (degree varchar(20)) inherits people; create table teachers (salary integer) inherits people; create table people of Person; create table students of Student under people; create table teachers of Teacher under people; Database System Concepts - 7th Edition 8.20 ©Silberschatz, Korth and Sudarshan Reference Types ▪ Reference Types allow you to create relationships between tables where one column contains references (pointers) to objects in another table. ▪ Creating reference types create type Person (ID varchar(20) primary key, name varchar(20), address varchar(20)) ref from(ID); create table people of Person; create type Department ( dept_name varchar(20), head ref(Person) scope people); create table departments of Department insert into departments values ('CS', '12345’) System generated references can be retrieved using subqueries ▪ (select ref(p) from people as p where ID = '12345') ▪ Using references in path expressions select head->name, head->address from departments; Database System Concepts - 7th Edition 8.21 ©Silberschatz, Korth and Sudarshan Object-Relational Mapping ▪ Object-relational mapping (ORM) systems allow Specification of mapping between programming language objects and database tuples Automatic creation of database tuples upon creation of objects Automatic update/delete of database tuples when objects are update/deleted Interface to retrieve objects satisfying specified conditions ▪ Tuples in database are queried, and object created from the tuples ▪ Details in Section 9.6.2 Hibernate ORM for Java Django ORM for Python Database System Concepts - 7th Edition 8.22 ©Silberschatz, Korth and Sudarshan Textual Data ▪ Information retrieval: querying of unstructured data Simple model of keyword queries: given query keywords, retrieve documents containing all the keywords More advanced models rank relevance of documents Today, keyword queries return many types of information as answers ▪ E.g., a query “cricket” typically returns information about ongoing cricket matches ▪ Relevance ranking : ▪ process of ordering and prioritizing search results or information based on their relevance to a user's query or request. Essential since there are usually many documents matching keywords Database System Concepts - 7th Edition 8.23 ©Silberschatz, Korth and Sudarshan Ranking using TF-IDF ▪ Term: keyword occurring in a document/query ▪ Term Frequency: TF(d, t), the relevance of a term t to a document d One definition: TF(d, t) = log(1 + n(d,t)/n(d)) where ▪ n(d,t) = number of occurrences of term t in document d ▪ n(d) = number of terms in document d ▪ Inverse document frequency: IDF(t) One definition: IDF(t) = 1/n(t) ▪ Relevance of a document d to a set of terms Q One definition: r(d, Q) = ∑t∈Q TF(d, t) ∗ IDF(t) Other definitions ▪ take proximity of words into account ▪ Stop words are often ignored Database System Concepts - 7th Edition 8.24 ©Silberschatz, Korth and Sudarshan Ranking Using Hyperlinks ▪ Hyperlinks provide very important clues to importance ▪ Google introduced PageRank, a measure of popularity/importance based on hyperlinks to pages Pages hyperlinked from many pages should have higher PageRank Pages hyperlinked from pages with higher PageRank should have higher PageRank Formalized by random walk model ▪ Let T[i, j] be the probability that a random walker who is on page i will click on the link to page j Assuming all links are equal, T[i, j] = 1∕Ni ▪ Then PageRank[j] for each page j can be defined as P[j] = δ∕N + (1 - δ) ∗ ∑i=1N (T[i, j] ∗ P[i]) Where N = total number of pages, and δ a constant usually set to 0.15 Database System Concepts - 7th Edition 8.25 ©Silberschatz, Korth and Sudarshan Ranking Using Hyperlinks ▪ Definition of PageRank is circular, but can be solved as a set of linear equations Simple iterative technique works well Initialize all P[i] = 1/N In each iteration use equation P[j] = δ∕N + (1 - δ) ∗ ∑i=1N (T[i, j] ∗ P[i]) to update P Stop iteration when changes are small, or some limit (say 30 iterations) is reached. ▪ Other measures of relevance are also important. For example: Keywords in anchor text Number of times who ask a query click on a link if it is returned as an answer Database System Concepts - 7th Edition 8.26 ©Silberschatz, Korth and Sudarshan Retrieval Effectiveness ▪ Measures of effectiveness Precision: what percentage of returned results are actually relevant Recall: what percentage of relevant results were returned At some number of answers, e.g. precision@10, recall@10 ▪ Keyword querying on structured data and knowledge bases Useful if users don’t know schema, or there is no predefined schema Can represent data as graphs Keywords match tuples Keyword search returns closely connected tuples that contain keywords ▪ E.g. on our university database given query “Zhang Katz”, Zhang matches a student, Katz an instructor and advisor relationship links them Database System Concepts - 7th Edition 8.27 ©Silberschatz, Korth and Sudarshan Spatial Data Database System Concepts - 7th Edition 8.28 ©Silberschatz, Korth and Sudarshan Spatial Data ▪ Spatial databases store information related to spatial locations, and support efficient storage, indexing and querying of spatial data. Geographic data -- road maps, land-usage maps, topographic elevation maps, political maps showing boundaries, land-ownership maps, and so on. ▪ Geographic information systems are special-purpose databases tailored for storing geographic data. ▪ Round-earth coordinate system may be used (Latitude, longitude, elevation) Geometric data: design information about how objects are constructed. For example, designs of buildings, aircraft, layouts of integrated- circuits. ▪ 2 or 3 dimensional Euclidean space with (X, Y, Z) coordinates Database System Concepts - 7th Edition 8.29 ©Silberschatz, Korth and Sudarshan Represented of Geometric Information Various geometric constructs can be represented in a database in a normalized fashion (see next slide) ▪ A line segment can be represented by the coordinates of its endpoints. ▪ A polyline or linestring consists of a connected sequence of line segments and can be represented by a list containing the coordinates of the endpoints of the segments, in sequence. Approximate a curve by partitioning it into a sequence of segments ▪ Useful for two-dimensional features such as roads. ▪ Some systems also support circular arcs as primitives, allowing curves to be represented as sequences of arc ▪ Polygons is represented by a list of vertices in order. The list of vertices specifies the boundary of a polygonal region. Can also be represented as a set of triangles (triangulation) Database System Concepts - 7th Edition 8.30 ©Silberschatz, Korth and Sudarshan Representation of Geometric Constructs Database System Concepts - 7th Edition 8.31 ©Silberschatz, Korth and Sudarshan Representation of Geometric Information (Cont.) ▪ Representation of points and line segment in 3-D similar to 2-D, except that points have an extra z component ▪ Represent arbitrary polyhedra by dividing them into tetrahedrons, like triangulating polygons. ▪ Alternative: List their faces, each of which is a polygon, along with an indication of which side of the face is inside the polyhedron. ▪ Geometry and geography data types supported by many databases E.g. SQL Server and PostGIS point, linestring, curve, polygons Collections: multipoint, multilinestring, multicurve, multipolygon LINESTRING(1 1, 2 3, 4 4) POLYGON((1 1, 2 3, 4 4, 1 1)) Type conversions: ST GeometryFromText() and ST GeographyFromText() Operations: ST Union(), ST Intersection(), … Database System Concepts - 7th Edition 8.32 ©Silberschatz, Korth and Sudarshan Design Databases ▪ Represent design components as objects (generally geometric objects); the connections between the objects indicate how the design is structured. ▪ Simple two-dimensional objects: points, lines, triangles, rectangles, polygons. ▪ Complex two-dimensional objects: formed from simple objects via union, intersection, and difference operations. ▪ Complex three-dimensional objects: formed from simpler objects such as spheres, cylinders, and cuboids, by union, intersection, and difference operations. ▪ Wireframe models represent three-dimensional surfaces as a set of simpler objects. Database System Concepts - 7th Edition 8.33 ©Silberschatz, Korth and Sudarshan Representation of Geometric Constructs ▪ Design databases also store non-spatial information about objects (e.g., construction material, color, etc.) ▪ Spatial integrity constraints are important. E.g., pipes should not intersect, wires should not be too close to each other, etc. Database System Concepts - 7th Edition 8.34 ©Silberschatz, Korth and Sudarshan Geographic Data ▪ Raster data consist of bit maps or pixel maps, in two or more dimensions. Example 2-D raster image: satellite image of cloud cover, where each pixel stores the cloud visibility in a particular area. Additional dimensions might include the temperature at different altitudes at different regions, or measurements taken at different points in time. ▪ Design databases generally do not store raster data. Database System Concepts - 7th Edition 8.35 ©Silberschatz, Korth and Sudarshan Geographic Data (Cont.) ▪ Vector data are constructed from basic geometric objects: points, line segments, triangles, and other polygons in two dimensions, and cylinders, spheres, cuboids, and other polyhedrons in three dimensions. ▪ Vector format often used to represent map data. Roads can be considered as two-dimensional and represented by lines and curves. Some features, such as rivers, may be represented either as complex curves or as complex polygons, depending on whether their width is relevant. Features such as regions and lakes can be depicted as polygons. Database System Concepts - 7th Edition 8.36 ©Silberschatz, Korth and Sudarshan Spatial Queries ▪ Region queries deal with spatial regions. e.g., ask for objects that lie partially or fully inside a specified region E.g., PostGIS ST_Contains(), ST_Overlaps(), … ▪ Nearness queries request objects that lie near a specified location. ▪ Nearest neighbor queries, given a point or an object, find the nearest object that satisfies given conditions. ▪ Spatial graph queries request information based on spatial graphs E.g., shortest path between two points via a road network ▪ Spatial join of two spatial relations with the location playing the role of join attribute. ▪ Queries that compute intersections or unions of regions Database System Concepts - 7th Edition 8.37 ©Silberschatz, Korth and Sudarshan End of Chapter 8 Database System Concepts - 7th Edition 8.38 ©Silberschatz, Korth and Sudarshan