This action might not be possible to undo. Are you sure you want to continue?

# Data Mining: Concepts and Techniques

**— Slides for Textbook — — Chapter 2 —
**

©Jiawei Han and Micheline Kamber Intelligent Database Systems Research Lab School of Computing Science Simon Fraser University, Canada http://www.cs.sfu.ca

September 14, 2010 Data Mining: Concepts and Techniques 1

**Chapter 2: Data Warehousing and OLAP Technology for Data Mining
**

What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining

Data Mining: Concepts and Techniques 2

September 14, 2010

**What is Data Warehouse?
**

Defined in many different ways, but not rigorously. A decision support database that is maintained separately from the organization’s operational database Support information processing by providing a solid platform of consolidated, historical data for analysis. “A data warehouse is a subject-oriented, integrated, time-variant, and nonvolatile collection of data in support of management’s decision-making process.”—W. H. Inmon Data warehousing: The process of constructing and using data warehouses

September 14, 2010

Data Mining: Concepts and Techniques

3

such as customer. 2010 Data Mining: Concepts and Techniques 4 . sales. Provide a simple and concise view around particular subject issues by excluding data that are not useful in the decision support process.Data Warehouse—SubjectOriented Organized around major subjects. September 14. product. not on daily operations or transaction processing. Focusing on the modeling and analysis of data for decision makers.

Data Warehouse—Integrated Constructed by integrating multiple. etc.g.. flat files. it is converted. breakfast covered. etc. 2010 . heterogeneous data sources relational databases. encoding structures. tax. Ensure consistency in naming conventions. When data is moved to the warehouse. Hotel price: currency. among different data sources E. on-line transaction records Data cleaning and data integration techniques are applied. attribute measures. Data Mining: Concepts and Techniques 5 September 14.

Every key structure in the data warehouse September 14.g. Data warehouse data: provide information from a historical perspective (e. Operational database: current value data. explicitly or implicitly But the key of operational data may or may not contain “time element”. 2010 Data Mining: Concepts and Techniques 6 .Data Warehouse—Time Variant The time horizon for the data warehouse is significantly longer than that of operational systems.. past 5-10 years) Contains an element of time.

2010 Data Mining: Concepts and Techniques 7 . and concurrency control mechanisms Requires only two operations in data accessing: initial loading of data and access of data. September 14.Data Warehouse—Non-Volatile A physically separate store of data transformed from the operational environment. Does not require transaction processing. recovery. Operational update of data does not occur in the data warehouse environment.

**Data Warehouse vs. Heterogeneous DBMS
**

**Traditional heterogeneous DB integration:
**

**Build wrappers/mediators on top of heterogeneous databases Query driven approach
**

When a query is posed to a client site, a meta-dictionary is used to translate the query into queries appropriate for individual heterogeneous sites involved, and the results are integrated into a global answer set Complex information filtering, compete for resources

**Data warehouse: update-driven, high performance
**

Information from heterogeneous sources is integrated in advance and stored in warehouses for direct query and analysis

September 14, 2010

Data Mining: Concepts and Techniques

8

**Data Warehouse vs. Operational DBMS
**

**OLTP (on-line transaction processing)
**

Major task of traditional relational DBMS Day-to-day operations: purchasing, inventory, banking, manufacturing, payroll, registration, accounting, etc. Major task of data warehouse system Data analysis and decision making User and system orientation: customer vs. market Data contents: current, detailed vs. historical, consolidated Database design: ER + application vs. star + subject View: current, local vs. evolutionary, integrated Access patterns: update vs. read-only but complex queries

**OLAP (on-line analytical processing)
**

**Distinct features (OLTP vs. OLAP):
**

September 14, 2010

Data Mining: Concepts and Techniques

9

OLTP vs. OLAP

OLTP users function DB design data usage access unit of work # records accessed #users DB size metric

September 14, 2010

OLAP knowledge worker decision support subject-oriented historical, summarized, multidimensional integrated, consolidated ad-hoc lots of scans complex query millions hundreds 100GB-TB query throughput, response

10

clerk, IT professional day to day operations application-oriented current, up-to-date detailed, flat relational isolated repetitive read/write index/hash on prim. key short, simple transaction tens thousands 100MB-GB transaction throughput

Data Mining: Concepts and Techniques

Why Separate Data Warehouse? High performance for both systems DBMS— tuned for OLTP: access methods. recovery Warehouse—tuned for OLAP: complex OLAP queries. concurrency control. consolidation. codes and formats which have to be reconciled Data Mining: Concepts and Techniques 11 September 14. multidimensional view. Different functions and different data: missing data: Decision support requires historical data which operational DBs do not typically maintain data consolidation: DS requires consolidation (aggregation. indexing. summarization) of data from heterogeneous sources data quality: different sources typically use inconsistent data representations. 2010 .

Chapter 2: Data Warehousing and OLAP Technology for Data Mining What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining Data Mining: Concepts and Techniques 12 September 14. 2010 .

Data Mining: Concepts and Techniques 13 September 14. brand.From Tables and Spreadsheets to Data Cubes A data warehouse is based on a multidimensional data model which views data in the form of a data cube A data cube. The top most 0-D cuboid. an n-D base cube is called a base cuboid. allows data to be modeled and viewed in multiple dimensions Dimension tables. such as item (item_name. such as sales. which holds the highest-level of summarization. is called the apex cuboid. quarter. type). week. or time(day. year) Fact table contains measures (such as dollars_sold) and keys to each of the related dimension tables In data warehousing literature. 2010 . The lattice of cuboids forms a data cube. month.

supplier time.supplier time.location item.supplier time.supplier 2-D cuboids 3-D cuboids 4-D(base) cuboid time.supplier item.supplier time.item.location location.Cube: A Lattice of Cuboids all time item location supplier 0-D(apex) cuboid 1-D cuboids time. supplier September 14.location item.location.location. item. 2010 Data Mining: Concepts and Techniques 14 . location.item.item time.

Conceptual Modeling of Data Warehouses Modeling data warehouses: dimensions & measures Star schema: A fact table in the middle connected to a set of dimension tables Snowflake schema: A refinement of star schema where some dimensional hierarchy is normalized into a set of smaller dimension tables. forming a shape similar to snowflake Fact constellations: Multiple fact tables share dimension tables. 2010 Data Mining: Concepts and Techniques 15 . therefore called galaxy schema or fact constellation September 14. viewed as a collection of stars.

2010 Data Mining: Concepts and Techniques 16 .Example of Star Schema time time_key day day_of_the_week month quarter year item Sales Fact Table time_key item_key branch_key location_key units_sold dollars_sold avg_sales Measures item_key item_name brand type supplier_type branch branch_key branch_name branch_type location location_key street city province_or_street country September 14.

2010 Data Mining: Concepts and Techniques .Example of Snowflake Schema time time_key day day_of_the_week month quarter year item Sales Fact Table time_key item_key branch_key item_key item_name brand type supplier_key supplier supplier_key supplier_type branch branch_key branch_name branch_type location_key units_sold dollars_sold avg_sales Measures location location_key street city_key city city_key city province_or_street country 17 September 14.

2010 Data Mining: Concepts and Techniques .Example of Fact Constellation time time_key day day_of_the_week month quarter year item Sales Fact Table time_key item_key branch_key item_key item_name brand type supplier_type Shipping Fact Table time_key item_key shipper_key from_location branch branch_key branch_name branch_type location_key units_sold dollars_sold avg_sales Measures location location_key street city province_or_street country to_location dollars_cost units_shipped shipper shipper_key shipper_name location_key shipper_type 18 September 14.

A Data Mining Query Language. DMQL: Language Primitives Cube Definition (Fact Table) define cube <cube_name> [<dimension_list>]: <measure_list> Dimension Definition ( Dimension Table ) define dimension <dimension_name> as (<attribute_or_subdimension_list>) Special Case (Shared Dimension Tables) First time as “cube definition” define dimension <dimension_name> as <dimension_name_first_time> in cube <cube_name_first_time> September 14. 2010 Data Mining: Concepts and Techniques 19 .

day_of_week. branch. street. quarter. province_or_state. city.Defining a Star Schema in DMQL define cube sales_star [time. location]: dollars_sold = sum(sales_in_dollars). supplier_type) define dimension branch as (branch_key. branch_name. branch_type) define dimension location as (location_key. 2010 Data Mining: Concepts and Techniques 20 . item_name. brand. month. country) September 14. item. day. units_sold = count(*) define dimension time as (time_key. avg_sales = avg(sales_in_dollars). year) define dimension item as (item_key. type.

year) define dimension item as (item_key. branch. location]: dollars_sold = sum(sales_in_dollars). province_or_state. day. city(city_key. branch_type) define dimension location as (location_key. type. item_name. units_sold = count(*) define dimension time as (time_key. day_of_week.Defining a Snowflake Schema in DMQL define cube sales_snowflake [time. country)) September 14. supplier(supplier_key. brand. street. item. month. branch_name. quarter. 2010 Data Mining: Concepts and Techniques 21 . supplier_type)) define dimension branch as (branch_key. avg_sales = avg(sales_in_dollars).

shipper_name. quarter. shipper. city. type. item. day_of_week. brand. units_sold = count(*) define dimension time as (time_key. from_location. item. branch. country) define cube shipping [time. avg_sales = avg(sales_in_dollars). location as location in cube sales. unit_shipped = count(*) define dimension time as time in cube sales define dimension item as item in cube sales define dimension shipper as (shipper_key. 2010 Data Mining: Concepts and Techniques 22 .Defining a Fact Constellation in DMQL define cube sales [time. branch_name. day. to_location]: dollar_cost = sum(cost_in_dollars). shipper_type) define dimension from_location as location in cube sales define dimension to_location as location in cube sales September 14. province_or_state. month. supplier_type) define dimension branch as (branch_key. branch_type) define dimension location as (location_key. street. item_name. location]: dollars_sold = sum(sales_in_dollars). year) define dimension item as (item_key.

sum().g. E. algebraic: if it can be computed by an algebraic function with M arguments (where M is a bounded integer). September 14.. min_N(). rank(). min().g. holistic: if there is no constant bound on the storage size needed to describe a subaggregate. E. median(). max().Measures: Three Categories distributive: if the result derived by applying the function to n aggregate values is the same as that derived by applying the function on all the data without partitioning.. standard_deviation(). avg().g. mode(). E. each of which is obtained by applying a distributive aggregate function. count().. 2010 Data Mining: Concepts and Techniques 23 .

Wind 24 Data Mining: Concepts and Techniques ... North_America Canada . Mexico Frankfurt . M..A Concept Hierarchy: Dimension (location) all region country city office September 14.. L... Spain . Chan Toronto ....... Vancouver . 2010 all Europe Germany .

. week} < year Set_grouping hierarchy {1. 2010 Data Mining: Concepts and Techniques 25 .View of Warehouses and Hierarchies Specification of hierarchies Schema hierarchy day < {month < quarter.10} < inexpensive September 14.

month.Multidimensional Data Sales volume as a function of product. and region Industry Region Re gi on Dimensions: Product. Location. 2010 Data Mining: Concepts and Techniques 26 . Time Hierarchical summarization paths Year Category Country Quarter Product Product City Office Month Day Week Month September 14.

A Canada Mexico sum September 14. U.S.A.A Sample Data Cube Pr od uc t TV PC VCR sum 1Qtr 2Qtr Date 3Qtr 4Qtr sum Total annual sales of TV in U. 2010 Data Mining: Concepts and Techniques Country 27 .S.

country 1-D cuboids 2-D cuboids 3-D(base) cuboid product. date. country September 14.Cuboids Corresponding to the Cube all 0-D(apex) cuboid product product.date date product.country country date. 2010 Data Mining: Concepts and Techniques 28 .

Browsing a Data Cube Visualization OLAP capabilities Interactive manipulation 29 September 14. 2010 Data Mining: Concepts and Techniques .

Typical OLAP Operations Roll up (drill-up): summarize data by climbing up hierarchy or by dimension reduction from higher level summary to lower level summary or detailed data. or introducing new dimensions project and select reorient the cube. visualization. 3D to series of 2D planes. 2010 . drill across: involving (across) more than one fact table drill through: through the bottom level of the cube to its back-end relational tables (using SQL) Data Mining: Concepts and Techniques 30 Drill down (roll down): reverse of roll-up Slice and dice: Pivot (rotate): Other operations September 14.

2010 Data Mining: Concepts and Techniques footprint Location .A Star-Net Query Model Shipping Method Customer Orders Customer CONTRACTS ORDER PRODUCT LINE ANNUALY QTRLY CITY COUNTRY DISTRICT REGION DIVISION Organization 31 AIR-EXPRESS TRUCK Time DAILY Product PRODUCT ITEM PRODUCT GROUP SALES PERSON Each circle is Promotion called a September 14.

2010 .Chapter 2: Data Warehousing and OLAP Technology for Data Mining What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining Data Mining: Concepts and Techniques 32 September 14.

2010 . and managed by operational systems consists of fact tables and dimension tables sees the perspectives of data in the warehouse from the view of end-user Data Mining: Concepts and Techniques 33 Data source view Data warehouse view Business query view September 14. stored.Design of a Data Warehouse: A Business Analysis Framework Four views regarding the design of a data warehouse Top-down view allows selection of the relevant information necessary for the data warehouse exposes the information being captured.

Data Warehouse Design Process Top-down.g. orders. bottom-up approaches or a combination of both Top-down: Starts with overall design and planning (mature) Bottom-up: Starts with experiments and prototypes (rapid) From software engineering point of view Waterfall: structured and systematic analysis at each step before proceeding to the next Spiral: rapid generation of increasingly functional systems.. e. Choose the grain (atomic level of data) of the business process Choose the dimensions that will apply to each fact table record Choose the measure that will populate each fact table record September 14. 2010 Data Mining: Concepts and Techniques 34 . short turn around time. quick turn around Typical data warehouse design process Choose a business process to model. invoices. etc.

2010 Data Storage OLAP Engine Front-End Tools 35 Data Mining: Concepts and Techniques .Multi-Tiered Architecture other Metadata source s Operational Monitor & Integrator OLAP Server DBs Extract Transform Load Refresh Data Warehouse Serve Analysis Query Reports Data mining Data Marts Data Sources September 14.

Three Data Warehouse Models Enterprise warehouse collects all of the information about subjects spanning the entire organization Data Mart a subset of corporate-wide data that is of value to a specific groups of users. dependent (directly from warehouse) data mart Virtual warehouse A set of views over operational databases Only some of the possible summary views may be materialized September 14. 2010 Data Mining: Concepts and Techniques 36 . selected groups. Its scope is confined to specific. such as marketing data mart Independent vs.

Data Warehouse Development: A Recommended Approach Distributed Data Marts Multi-Tier Data Warehouse Data Mart Data Mart Model refinement Enterprise Data Warehouse Model refinement Define a high-level corporate data model September 14. 2010 Data Mining: Concepts and Techniques 37 .

. low level: relational. high-level: array Specialized SQL servers specialized support for SQL queries over star/snowflake schemas September 14. e.g. and additional tools and services greater scalability Multidimensional OLAP (MOLAP) Array-based multidimensional storage engine (sparse matrix techniques) fast indexing to pre-computed summarized data Hybrid OLAP (HOLAP) User flexibility. implementation of aggregation navigation logic.OLAP Server Architectures Relational OLAP (ROLAP) Use relational or extended-relational DBMS to store and manage warehouse data and OLAP middle ware to support missing pieces Include optimization of DBMS backend. 2010 Data Mining: Concepts and Techniques 38 .

2010 .Chapter 2: Data Warehousing and OLAP Technology for Data Mining What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining Data Mining: Concepts and Techniques 39 September 14.

September 14. 2010 Data Mining: Concepts and Techniques 40 . or some (partial materialization) Selection of which cuboids to materialize Based on size. etc. sharing. access frequency.Efficient Data Cube Computation Data cube can be viewed as a lattice of cuboids The bottom-most cuboid is the base cuboid The top-most cuboid (apex) contains only one cell How many cuboids in an n-dimensional cube with L levels? Materialization = ∏ ( L +1) of n data cube T i i =1 (cuboid) (full materialization). none (no Materialize every materialization).

’96) SELECT item. customer). city. year) Data Mining: Concepts and Techniques 41 (item) (year) September 14. customer). (customer) () (city. city. year]: sum(sales_in_dollars) compute cube sales Transform it into a SQL-like language (with a new operator cube by. item. (city.Cube Operation Cube definition and computation in DMQL define cube sales[item. (product). product. item) (city. year) (date. city. (product. 2010 . year Need compute the following Group-Bys (date. (date).(date. SUM (amount) FROM SALES (city) () CUBE BY item.product). year. customer). year) (item. introduced by Gray et al.

Cube Computation: ROLAP-Based Method Efficient cube computation methods ROLAP-based cubing algorithms (Agarwal et al’96) Array-based cubing algorithm (Zhao et al’97) Bottom-up computation method (Bayer & Ramarkrishnan’99) ROLAP-based cubing algorithms Sorting. rather than from the base fact table September 14. and grouping operations are applied to the dimension attributes in order to reorder and cluster related tuples Grouping is performed on some subaggregates as a “partial grouping step” Aggregates may be computed from previously computed aggregates. hashing. 2010 Data Mining: Concepts and Techniques 42 .

Cache-results: caching results of a cuboid from which other cuboids are computed to reduce disk I/Os Amortize-scans: computing as many as possible cuboids at the same time to amortize disk reads Share-sorts: sharing sorting costs cross multiple cuboids when sort-based method is used Share-partitions: sharing the partitioning cost cross multiple cuboids when hash-based algorithms are used September 14. VLDB’96) Smallest-parent: computing a cuboid from the smallest cubod previously computed cuboid. 2010 Data Mining: Concepts and Techniques 43 . al.Cube Computation: ROLAP-Based Method (2) This is not in the textbook but in a research paper Hash/sort based methods (Agarwal et.

offset) Compute aggregates in “multiway” by visiting cube cells in the order which minimizes the # of times to visit each cell.Multi-way Array Aggregation for Cube Computation Partition arrays into chunks (a small subcube which fits in memory). Compressed sparse array addressing: (chunk_id. and reduces memory access and storage cost. 2010 A Data Mining: Concepts and Techniques 44 . C c3 61 62 63 64 c2 45 46 47 48 c1 29 30 31 32 c0 b3 B 13 9 5 1 a0 14 15 16 B b2 b1 b0 2 a1 3 a2 4 a3 60 44 28 56 40 24 52 36 20 What is the best traversing order to do multi-way aggregation? September 14.

2010 Data Mining: Concepts and Techniques 45 .Multi-way Array Aggregation for Cube Computation C c3 61 62 63 64 c2 45 46 47 48 c1 29 30 31 32 c0 B 13 9 5 1 a0 2 a1 3 a2 4 a3 14 15 16 28 24 20 44 40 36 60 56 52 b3 B b2 b1 b0 A September 14.

2010 Data Mining: Concepts and Techniques 46 .Multi-way Array Aggregation for Cube Computation C c3 61 62 63 64 c2 45 46 47 48 c1 29 30 31 32 c0 B 13 9 5 1 a0 2 a1 3 a2 4 a3 14 15 16 28 24 20 44 40 36 60 56 52 b3 B b2 b1 b0 A September 14.

12 (pp. See the details of Example 2.) Method: the planes should be sorted and computed according to their size in ascending order. fetch and compute only one chunk at a time for the largest plane Limitation of the method: computing well only for a small number of dimensions If there are a large number of dimensions. 2010 Data Mining: Concepts and Techniques 47 . 75-78) Idea: keep the smallest plane in the main memory. “bottomup computation” and iceberg cube computation methods can be explored September 14.Multi-Way Array Aggregation for Cube Computation (Cont.

Indexing OLAP Data: Bitmap Index Index on a particular column Each value in the column has a bit vector: bit-op is fast The length of the bit vector: # of records in the base table The i-th bit is set if the i-th row of the base table has the value for the indexed column not suitable for high cardinality domains Base table Cust C1 C2 C3 C4 C5 Index on Region Index on Type RecID Retail Dealer 1 1 0 2 0 1 3 0 1 4 1 0 5 0 1 48 Region Type RecID Asia Europe America 1 1 0 0 Asia Retail 0 1 0 Europe Dealer 2 3 1 0 0 Asia Dealer 4 0 0 1 America Retail 5 0 1 0 Europe Dealer Data Mining: Concepts and Techniques September 14. 2010 .

S-id) where R (R-id. …) S (S-id. 2010 .g. fact table: Sales and two dimensions city and product A join index on city maintains for each distinct city a list of R-IDs of the tuples recording the Sales in the city Join indices can span multiple dimensions Data Mining: Concepts and Techniques 49 September 14. E. join index relates the values of the dimensions of a start schema to rows in the fact table. …) Traditional indices map the values to a list of record ids It materializes relational join in JI file and speeds up relational join — a rather costly operation In data warehouses.Indexing OLAP Data: Join Indices Join index: JI(R-id.

2010 Data Mining: Concepts and Techniques 50 . dice = selection + projection Determine to which materialized cuboid(s) the relevant operations should be applied. roll.Efficient Processing OLAP Queries Determine which operations should be performed on the available cuboids: transform drill. e. Exploring indexing structures and compressed vs.g. into corresponding SQL and/or OLAP operations. dense array structures in MOLAP September 14. etc.

currency of data (active. view.Metadata Repository Meta data is the data defining warehouse objects. hierarchies. or purged). derived data defn. dimensions. It has the following kinds Description of the structure of the warehouse schema. audit trails) Operational meta-data The algorithms used for summarization The mapping from operational environment to the data warehouse Data related to system performance warehouse schema. 2010 Data Mining: Concepts and Techniques 51 . monitoring information (warehouse usage statistics. error reports. ownership of data. data mart locations and contents data lineage (history of migrated data and transformation path). view and derived data definitions business terms and definitions. charging policies Business data September 14. archived.

and build indicies and partitions propagate the updates from the data sources to the warehouse Data Mining: Concepts and Techniques 52 Data cleaning: Data transformation: Load: Refresh September 14. heterogeneous. compute views. summarize. check integrity. consolidate. and external sources detect errors in the data and rectify them when possible convert data from legacy or host format to warehouse format sort.Data Warehouse Back-End Tools and Utilities Data extraction: get data from multiple. 2010 .

Chapter 2: Data Warehousing and OLAP Technology for Data Mining What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining Data Mining: Concepts and Techniques 53 September 14. 2010 .

InExp. 2010 Data Mining: Concepts and Techniques 54 .’98) pre-compute measures indicating exceptions. huge search space Discovery-driven (Sarawagi et al. guide user in the data analysis. at all levels of aggregation Exception: significantly different from the value anticipated. and PathExp values) can be overlapped with cube construction September 14. based on a statistical model Visual cues such as background color are used to reflect the degree of exception of each cell Computation of exception indicator (modeling fitting and computing SelfExp.Discovery-Driven Exploration of Data Cubes Hypothesis-driven: exploration by user.

2010 Data Mining: Concepts and Techniques 55 .Examples: Discovery-Driven Data Cubes September 14.

region. month: R such that R. et al. region. 1998): Compute complex queries involving multiple dependent aggregates at multiple granularities Ex. and the total sales among all maximum price tuples select item. month. region. Grouping by all subsets of {item. 2010 Data Mining: Concepts and Techniques 56 . sum(R. find the maximum price in 1997 for each group.sales) from purchases where year = 1997 cube by item. find the min and max shelf life. max(price). and find the fraction of the total sales due to tuple that have min shelf life within the set of all max price tuples September 14.Complex Aggregation at Multiple Granularities: Multi-Feature Cubes Multi-feature cubes (Ross. among the max price tuples.price = max(price) Continuing the last example. month}.

Chapter 2: Data Warehousing and OLAP Technology for Data Mining What is a data warehouse? A multi-dimensional data model Data warehouse architecture Data warehouse implementation Further development of data cube technology From data warehousing to data mining Data Mining: Concepts and Techniques 57 September 14. 2010 .

pivoting knowledge discovery from hidden patterns supports associations. and reporting using crosstabs.Data Warehouse Usage Three kinds of data warehouse applications Information processing supports querying. basic statistical analysis. Analytical processing Data mining Differences among the three tasks Data Mining: Concepts and Techniques 58 September 14. and presenting the mining results using visualization tools. charts and graphs multidimensional analysis of data warehouse data supports basic OLAP operations. 2010 . performing classification and prediction. constructing analytical models. drilling. slice-dice. tables.

From On-Line Analytical Processing to On Line Analytical Mining (OLAM) Why online analytical mining? High quality of data in data warehouses DW contains integrated. dicing. Architecture of OLAM Data Mining: Concepts and Techniques 59 September 14. consistent. cleaned data Available information processing structure surrounding data warehouses ODBC. 2010 . OLEDB. Web accessing. pivoting. service facilities. algorithms. etc. and tasks. On-line selection of data mining functions integration and swapping of multiple mining functions. reporting and OLAP tools OLAP-based exploratory data analysis mining with drilling.

2010 Data Data integration Warehouse Data Mining: Concepts and Techniques Data Repository 60 .An OLAM Architecture Mining query User GUI API Mining result Layer4 User Interface OLAM Engine Data Cube API OLAP Engine Layer3 OLAP/OLAM MDDB Database API Data cleaning Layer2 MDDB Meta Data Filtering Filtering&Integration Layer1 Databases September 14.

2010 . slicing.Summary Data warehouse A subject-oriented. HOLAP Efficient computation of data cubes Partial vs. time-variant. and nonvolatile collection of data in support of management’s decision-making process Star schema. dicing and pivoting OLAP servers: ROLAP. snowflake schema. rolling. MOLAP. fact constellations A data cube consists of dimensions & measures A multi-dimensional model of a data warehouse OLAP operations: drilling. no materialization Multiway array aggregation Bitmap index and join index implementations Discovery-drive and multi-feature cubes From OLAP to OLAM (on-line analytical mining) Data Mining: Concepts and Techniques 61 Further development of data cube technology September 14. full vs. integrated.

R. Agrawal. Ramakrishnan. Dayal. Gupta. Bombay. 1997 ACM-SIGMOD Int. In Proc. ACM SIGMOD Record. Data Mining and Knowledge Discovery. Sept. Ramakrishnan. K. Beyer and R. Birmingham. Conf. MDAPI specification version 2. Data cube: A relational aggregation operator generalizing group-by. Bosworth. June 1998. In Proc. Agarwal. Modeling multidimensional databases. In Proc. 359-370. M. J. F. Sarawagi. Gupta. 1998 ACM-SIGMOD Int. Conf. J. Seattle. In http://www. Raghavan. A. 1997 Int. Washington. cross-tab and subtotals. Management of Data (SIGMOD'99). Chaudhuri. Layman. Pellow. Very Large Data Bases. E. Gehrke. 506-521. Reichart. Philadelphia. A. Conf. 1998. Efficient view maintenance in data warehouses. An overview of data warehousing and OLAP technology. 417-427. Singh. Gunopulos. 232-243. Sarawagi. Tucson. 1:29-54. R. Automatic subspace clustering of high dimensional data for data mining applications. R. England. Agrawal. On the computation of multidimensional aggregates. J. Management of Data.htm. and S. In Proc.0. R. A. Pirahesh. Agrawal.olapcouncil. 1999 ACM-SIGMOD Int. A. and S. Data Mining: Concepts and Techniques 62 September 14. and H. 94-105. Arizona. F. D. M. D. In Proc. Chaudhuri and U.References (I) S. S. 1996. Naughton. April 1997. Agrawal. S. OLAP council. Conf. Bottom-Up Computation of Sparse and Iceberg CUBEs. P. Abbadi. June 1999. 26:65-74. Yurek.org/research/apily. A. May 1997. and T. 1997. Data Engineering. 1997. D. A. Conf. Deshpande. Gray. India. Management of Data. Venkatrao. 2010 . 1996 Int. PA. and P.

John Wiley & Sons. Spain. Zhao. Discovery-driven exploration of OLAP data cubes. and D. Naughton. and N. E. In Proc. of Extending Database Technology (EDBT'98). May 1997. Srivastava. pages 168-182. 1997 Int. Tucson. and J. Spain.microsoft. Int. Valencia. Sarawagi. March 1998. 1997.References (II) V.0. OLAP Solutions: Building Multidimensional Information Systems. An array-based algorithm for simultaneous multidimensional aggregates. 1998. Microsoft. of Extending Database Technology (EDBT'98). Conf. Int. Greece. Management of Data.com/data/oledb/olap. Athens. 2010 . R. Srivastava. June 1996. In Proc. Y. Aug. Thomsen. F. In Proc. D. Chatziantoniou. 116-125. Ross. Megiddo. Harinarayan. Complex aggregation at multiple granularities. Fast computation of sparse datacubes. In http://www. M. Conf. Montreal. 159-170. pages 205-216. Canada. March 1998. Valencia. A. In Proc. A. Agrawal. K. OLEDB for OLAP programmer's reference version 1. Data Mining: Concepts and Techniques 63 September 14. Conf. Management of Data. In Proc. Very Large Data Bases. Ullman. Conf. 1997. Deshpande. P. and J. Ross and D. S. 1996 ACM-SIGMOD Int. K. Rajaraman. D. Implementing data cubes efficiently. Conf. 263-277. Arizona. 1997 ACM-SIGMOD Int.

sfu. 2010 Data Mining: Concepts and Techniques 64 .cs.ca/~han Thank you !!! September 14.http://www.