Presentation is loading. Please wait.

Presentation is loading. Please wait.

Data and Knowledge Representation Lecture 6 Qing Zeng, Ph.D.

Similar presentations


Presentation on theme: "Data and Knowledge Representation Lecture 6 Qing Zeng, Ph.D."— Presentation transcript:

1 Data and Knowledge Representation Lecture 6 Qing Zeng, Ph.D.

2 Last Time We Talked About Medical Coding Systems Medical Coding Systems UMLS UMLS

3 Today We Will Talk About Major KR schemes Major KR schemes Semantic Network Semantic Network Frame-based Representation Frame-based Representation Production Rules Production Rules

4 Semantic Network An long existing notion: there are different pieces of knowledge of world, and they are all linked together through certain semantics. An long existing notion: there are different pieces of knowledge of world, and they are all linked together through certain semantics.

5 Basic Components Nodes Nodes Represent concepts Represent concepts Arcs Arcs Represent relations Represent relations Labels for nodes and arcs Labels for nodes and arcs

6 Little Constraint patient physician Nurse Interact

7 Little Constraint DSG Site Course Site Instructors’ Homepage Link Web

8 Relation Directed or non-directed Directed or non-directed Multiple relations between two concepts Multiple relations between two concepts Can have different properties Can have different properties Reflexive (e.g. concurrence) Reflexive (e.g. concurrence) Transitive (e.g. causal) Transitive (e.g. causal) Symmetric (e.g. sibling) Symmetric (e.g. sibling) ……….. ………..

9 Some Often Used Relations in Biomedical Domain IS A IS A IS PART OF IS PART OF CAUSE OF CAUSE OF MEASURES MEASURES CO-OCCURS CO-OCCURS ………… …………

10 Major Limitation Lack of Semantics Lack of Semantics No formal semantic of the relations No formal semantic of the relations E.g. Does “ISA” mean subclass, member, etc? E.g. Does “ISA” mean subclass, member, etc? Possible multiple interpretations Possible multiple interpretations Restricted expressiveness Restricted expressiveness E.g. can not distinguish between instance and class E.g. can not distinguish between instance and class

11 Extension Extending expressivity (distinguish different types of concepts and relations” Extending expressivity (distinguish different types of concepts and relations” Distinguish between “some” and “all” Distinguish between “some” and “all” Distinguish between “existence” and “intension” Distinguish between “existence” and “intension” Distinguish between “definition” and “assertion” Distinguish between “definition” and “assertion” Add semantic rigor Add semantic rigor Map to logic (Sowa – CG) Map to logic (Sowa – CG)

12 UMLS Semantic Network – Concept Hierarchy

13 UMLS Semantic Network – Relation Hierarchy

14 UMLS Semantic Network – Relation Constraints

15 Frame-based Network Distinguish instance vs. class Distinguish instance vs. class Hierarchical structure (superclass and subclass) Hierarchical structure (superclass and subclass) Multiple hierarchy Multiple hierarchy Slots Slots Member slot Member slot Own slot Own slot

16 Slot Frame identifying information Frame identifying information Relationship between frames Relationship between frames Descriptors of requirements for frame match Descriptors of requirements for frame match Procedural information Procedural information Default information Default information Restrictions and constraints Restrictions and constraints New instance information New instance information

17 Strength Help organize knowledge hierarchically Help organize knowledge hierarchically Procedure information Procedure information Support multiple inheritance Support multiple inheritance

18 Weakness Expressiveness (e.g. quantifier) Expressiveness (e.g. quantifier) Inheritance Inheritance Sub classing (override slot value) Sub classing (override slot value) Multiple inheritance Multiple inheritance Large complex knowledge system Large complex knowledge system

19 Example: MED

20 Example: Protégé

21

22

23 Production Rules Also called IF-THEN rules Also called IF-THEN rules Many forms: Many forms: IF condition THEN action IF condition THEN action IF premise THEN conclusion IF premise THEN conclusion IF proposition p1 and proposition p2 are true THEN proposition p3 is true IF proposition p1 and proposition p2 are true THEN proposition p3 is true

24 Components Rule base Rule base Inference engine Inference engine Working memory Working memory

25 Inference Modus ponens Modus ponens Forward chaining Forward chaining Modus tollens Modus tollens Background chaining Background chaining

26 Example: MYCIN IF the identity of the germ is not known with certainty AND the germ is gram-positive AND the morphology of the organism is "rod" AND the germ is aerobic THEN there is a strong probability (0.8) that the germ is of type enterobacteriacae

27 Pro and Con Pro Pro Modular Modular Natural Natural Con Con Not efficient Not efficient Not expressive Not expressive

28 Exercise “Mrs Z.N. : 35 yr old farmer labourer from the Kwazulu- Natal province. She presented with a one week history of right sided pleuritic chest pain. The onset was sudden. This was associated with a warm flushing feeling and dizziness. There was also a pain in the right upper quadrant of her abdomen. There was no history of shortness of breath, cough or wheezing. There were no cardiac symptoms, no symptoms of malaise or loss of weight. No history of fevers. She was not a smoker and drank no alcohol. Of note is that she admitted to eating the entrails of the goats that she keeps.” - http://www.wits.ac.za/fac/med/pulmonology/case1.htm

29 Exercise Represent the information in semantic network Represent the information in semantic network Represent the information in frame-based network Represent the information in frame-based network

30 Exercise The thyroid gland is located at the base of your neck in front of your trachea (or windpipe). It has two sides and is shaped like a butterfly. The thyroid gland is located at the base of your neck in front of your trachea (or windpipe). It has two sides and is shaped like a butterfly. The thyroid gland makes, stores, and releases two hormones - T4 (thyroxine) and T3 (triiodothyronine). Thyroid hormones control the rate at which every part of your body works. This is called your metabolism. Your metabolism controls whether you feel hot or cold or tired or rested. When your thyroid gland is working the way it should, your metabolism stays at a steady pace -not too fast or too slow. The thyroid gland makes, stores, and releases two hormones - T4 (thyroxine) and T3 (triiodothyronine). Thyroid hormones control the rate at which every part of your body works. This is called your metabolism. Your metabolism controls whether you feel hot or cold or tired or rested. When your thyroid gland is working the way it should, your metabolism stays at a steady pace -not too fast or too slow. If no cancer cells are found, your doctor may prescribe a thyroid hormone to decrease the size of your nodule. Or, your doctor may suggest surgery to remove it. If cancer cells are found, further treatment will be needed. Thyroid cancer usually can be treated with success. If no cancer cells are found, your doctor may prescribe a thyroid hormone to decrease the size of your nodule. Or, your doctor may suggest surgery to remove it. If cancer cells are found, further treatment will be needed. Thyroid cancer usually can be treated with success.

31 Excise Which representation scheme to choose? Which representation scheme to choose?

32 Knowledge Representation Process Identify Needs Identify Needs Conceptualization Conceptualization Formalization Formalization Implementation Implementation Evaluation Evaluation

33 Reading Sowa Chap. 4 Sowa Chap. 4


Download ppt "Data and Knowledge Representation Lecture 6 Qing Zeng, Ph.D."

Similar presentations


Ads by Google