Ontology Learning from Relational Database: Opportunities for Semantic Information Integration Flashcards

1
Q

What is the risk of traditional rule-based transformation & reverse engineering for constructing RBDs?

A

Semantic loss during transformation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Why use ontology?

A

Integrate heterogenous data from different info systems

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

NLP

A

Natural Language Processing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

OL

A

Ontology Learning

Create ontologies by extracting knowledge from text, dictionaries, & RBD by using techniques from data mining, ML, & NLP

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Name three recent top trends in Ontology Learning

A

relation extraction, ontology learning, and ontology mapping

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Why use Semantic Info Integration in ontology learning from RDB?

A

Address the bottlenecks of ontology-based integration & strengths of ontology learning form RDB in mapping results

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are obstacles to improving ontology quality?

A

Many tools are limited to one-to-one mapping or they are for small-scare ontologies
Or like info extracting rely on linguistic rules that need manual identification

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Which ontology learning model has the best accuracy: linguistics techniques, statistical techniques, or inductive logic programming

A

Inductive logic programming at 96%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are two categories & sub-categories of ontology learning approaches?

A

Linguistic-based & ML approach

ML broken down into sub-class: statistic-based approach & logic-based approach

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What are two critical phases of constructing an ontology from RDB?

A

1) Construct ontology from RDB schema

2) Generate an ontology instance from RDB data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

ERD

A

Entity Relationship Diagram

graph-based method to illustrate the entities, attributes, and their relationships at the conceptual level

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What process is used in RE?

A

Recursive process for rebuilding hierarchies from tables could recover the lost semantic information & database table during reverse engineering transformation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are three techniques for mapping?

A

Rule-based mapping, graph-based mapping, & similarity-based mapping

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

N = {Na;Nc;Ne}

A

finite set of node, while Na represents attribute node that is depicted as box, Nc represents class node depicted as ellipses, Ne represents event node that is depicted as triangles

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Name the three techniques of ontology learning from RDB?

A
Rule-based learning, semantic mining, & 
active learning (learn & extract knowledge from an unlabeled data set)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is a schema?

A

Database schema describes data.
Relational database schema structures a set of instances for efficient storage and querying. (The structure is specified as tables and columns)

17
Q

DBMS

A

Database mgmt system