(1) The paper presents SAMSTAR, a semi-automated method to generate star schemas from entity relationship diagrams (ERDs). (2) SAMSTAR uses the Connection Topology Value (CTV) to identify candidate fact tables based on entity relationships, and Annotated Dimensional Design Patterns (A_DDPs) to identify candidate dimension tables. (3) The algorithm calculates CTV scores for each entity to identify fact candidates, then uses direct and indirect relationships, WordNet synonyms, and A_DDP lists to identify dimension candidates for each fact.

1. SAMSTAR:of Poster
A Semi-Automated Lexical Method to
Titl
generate Star Schemas from an ERD
Authors
Ritu Khare, Il Yeol Song, Yuan An
PROBLEM
To develop a star schema, existing approaches analyze the
attributes of interesting business entities.
•Entities with numerical measure attributes are assumed to be the
candidates of facts
•Entities with non-numerical and descriptive attributes are assumed
to be the candidates of dimensions.
2. Annotated Dimension Design Patterns (A_DDP)
We have referred to four sources and have instantiated the six classes of
DDP to produce a list of 131 commonly used dimension entities. We refer
to these entities as Annotated DDP (A_DDPs). These entities are
frequently used entities in the business processes. Examples include
account, activity, agent, aircraft, airport, etc.
RESULTS
Invoice
We focus on the structure of the ERD. The novel features of
SAMSTAR are:
(1) the use of the notion of Connection Topology Value (CTV) in
identifying the candidates of facts and dimensions and
(2) the use of Annotated Dimensional Design Patterns (A_DDP) as
well as WordNet to extend the list of dimensions.
Customer
ALGORITHM
Quantitative
Method
(FIND FACTS
AND DIRECT
DIMENSIONS)
DDP
Fig.2 SAMSTAR Overview
Quantitative
Method
(FIND INDIRECT
DIMENSIONS
WordNet
A_DDP
1. Connection Topology Value (CTV)
CTV (e) 1* n 0.8 *
CTV(Node(i ))
i 1
* If you are scanning charts,
cartoons, illustrations or
plain text non-photo
represents scan entity having
images), an at 600 dpi,
then ‘Save As’ at 225 dpi.
where i
a direct M:1 relationship with e.
For Fig. 1, CTV is calculated in the following manner:
A
B
E
H
weight_d=100%; weight_i=80%
D
F
C
The CTV for each entity is:
CTV (H) = 1* 0 + 0.8 * 0 = 0
CTV(F) = 0
G
CTV(G) = 0
CTV(E) = 1*1 + 0.8 * CTV(H) = 1
Fig.1: Calculation of CTV
CTV(B)= 1*1 + 0.8 * (CTV(E))= 1.8
CTV(C)=1*2 + 0.8 * (CTV(G) + CTV( F)) = 2
CTV(D)=1*1 + 0.8 * (CTV(C)) = 1 + 0.8 * 2 = 2.6
CTV(A)= 1*2 + 0.8 * (CTV(B) + CTV( C)) = 2 + 0.8 * (1.8+2) = 5.04
www.ischool.drexel.edu
Order
Time
Logistics
Shipment
Order
Customer
Area
Shipment
Supplier
Product-Supplier
Order-Product
Promotion Store
Store
SAMSTAR
Order
Invoice
Customer
Store
Promotion
Store
Customer
Promotion Type
Product
Warehouse
Time
Star Schema
n
Return
Store
ER Diagram
OUR SOLUTION
Store
Invoice
Return
Warehouse
Hence, these approaches are qualitative in nature, and focus only
on the semantics of ERD.
Time
1. Pre-process the input ERD.
2. Store Entities and Relationships.
3. Let user choose weighting factors for direct and indirect relationships.
4. Calculate the CTV for all entities.
5. Calculate the threshold value, Th, for CTV.
6. Identify entities having CTV higher than the threshold Th.
These are the candidates for fact tables.
7. Decide and shortlist the fact entities.
8. For each fact entity, perform the following steps:
(i) Identify the entities having direct M:1 link with a fact entity.
(ii) Identify entities having indirect M:1 link with the fact entity.
Out of these entities, identify synonyms of entity names from WordNet.
Extract the terms which match the DDP entity list or the A_DDP List.
(iii) Combine the results to Steps 8(i) and 8(ii) to prepare a list of candidate
dimensions for a given fact.
(iv) Add time dimension to this list.
9. Decide the dimension entities.
10. Let the user post-process the Star Schemas
11. Generate the star schema(s).
Order
Product
Order
Product
Invoice
Fig.3: Result of SAMSTAR
Schemas generated by SAMSTAR are similar to those generated
by the manual steps in case study papers.
SAMSTAR generated star schemas are the superset of the ones
generated manually in the paper using the same ER diagrams,
user needs and business goals.
This shows our schemas are inclusive of all possible facts and
dimensions. This gives the designer a helpful aid to prune the
schema as per the business and user requirements.
CONTRIBUTION
•A universal method to generate star schema(s) in that we have
used generalized DDPs and WordNet to identify dimensions of a
fact table.
•Quantitative in nature in that we analyze the structure of an input
ERD.
•Identifies a set of fact candidates from a large and complex ERD.
•Automatable up to a large extent; simplifies the work of
experienced designers; and gives a smooth head-start to novices.
The paper SAMSTAR: A Semi-Automated Lexical Method for
generating Star Schemas using an Entity Relationship
Diagram was presented at Data warehousing and On-line
Analytical Processing (DOLAP 2007) workshop held in Lisbon,
Portugal, November 2007.