Chapter 3: Query Structures Matching Criteria

1
Chapter 3 Query Structures Matching Criteria

• Exact Match
• Numerical or business databases
• Range Match
• Numeric data like age, accounting, etc.
• Approximate Match
• Evaluation function

2
Method of Queries

• Boolean Queries
• Vector Queries
• Extended Boolean Queries
• Fuzzy Queries
• Probabilistic Queries
• Natural Language Queries

3
Boolean Queries

• Simplicity
• No good way to weight for significance
• Misstated query
• The order of precedence for the logical
  connectives
• Free to construct a highly complex query for user
• Controlling the size and composition of the
  retrieved set.
• Use Boolean algebra based on logical connectives
  such as AND, OR and NOT

4
Boolean Queries

• Disjunctive Normal Form (DNF)
• (A AND B) OR (A AND NOT C)
• Conjunctive Normal Form (CNF)
• (A OR B) AND (A OR NOT C)
• Not A OR (A OR B AND C)

5
Vector Queries

• Document is represented by a vector, or a list of
  terms
• Evaluation
• Queries match the terms by 0 or 1
• Weights can be given by system or user

6
Extended Boolean Queries

• Query Aw1 Bw2
• Case 1 denotes OR
• Case 2 denotes AND
• Case 3 denotes AND NOT

7
Extended Boolean Queries

• Rule for OR
• Compute set S1 B - A
• Compute the distance to the closest element of A
  in S1
• Keep the closest distance for the weighted
  portion of B in S1
• Compute set S2 A - B
• Compute the distance to the closest element of B
  in S2
• Keep the closest distance for the weighted
  portion of A in S2
• Compute set S3 A ? B
• Compute set S S1 ? S2 ? S3
• EX A 1,4,5,10,11,15,17,18,19,23
• B 1,2,4,5,7,8,10,13,17,22
• S A0.6 OR B0.4

8
Fuzzy Queries

• Use fuzzy set theory
• Uncertainty
• Membership grade ranging 0.0 to 1.0
• Example set of tall people

9
Probabilistic Queries

• Use probabilistic
• No uncertainty
• Example tossing of coin

10
Natural Language Queries

• The easiest for a user to formulate
• Imprecise
• Inaccurate
• Ungrammatical
• The most difficult for a developer to implement
• Syntactic
• Semantic
• Pragmatic

11
Matching Criteria

• Exact Match
• Boolean queries, Vector queries
• Range Match
• Numeric data like age, accounting, etc.
• Approximate Match
• Fuzzy, Probabilistic, Vector queries, Natural
  language

12
System Analysis

• Standardization of Communication
• Flowcharts
• Data Flow Diagram (DFD)
• Data Dictionaries
• Structured English
• Decision Tables
• Decision Trees
• Etc.

13
HTML Table Tags  
 
14
HTML practice

• ltbody bgcolor"pink"gt
• lttable border"2"gt lttrgt ltth colspan"10"gt
• lth1gtKi Tai's Profilelt/h1gt lt/thgt lt/trgt
• lttrgt lttd align"center" rowspan"6"gt ltimg
  src"file///e/ou5003fall/myimage.gif"gt lt/tdgt
  lt/trgt

15
HTML practice

• lttrgt lttd align"center"gt lth3gtKi Tai's
  Vitalslt/h3gt lt/tdgtlt/trgt
• lttrgtlttd align"center"gtHeight - 163 Cmlt/tdgtlt/trgt
• lttrgtlttd align"center"gtWeight - 58 Kglt/tdgtlt/trgt
• lttrgtlttd align"center"gtNationality -
  Koreanlt/tdgtlt/trgt
• lttrgtlttd align"center"gtHobby - Table
  Tennislt/tdgtlt/trgt lt/tablegt
• ltpgtThis is Ki Tai's personal history. If you
  want to send E-mail, click lta hrefmailtokjeong_at_o
  u.edugtKi Tai Jeonglt/agtlt/pgt

16
HTML practice

• lttable border"2"gtlttrgtltth colspan"3"
  bgcolor"green"gt2003lt/thgt
• lttrgtlttdgtJanuarylt/tdgtlttdgtFebruarylt/tdgtlttdgtMarchlt
  /tdgtlt/trgt
• lttrgtlttdgtAprillt/tdgtlttdgtMaylt/tdgtlttdgtJunelt/tdgtlt/t
  rgt

17
HTML practice

• lttrgtlttdgtJulylt/tdgtlttdgtAugustlt/tdgtlttdgtSeptemberlt/
  tdgtlt/trgt
• lttrgtlttdgtOctoberlt/tdgtlttdgtNovemberlt/tdgtlttdgtDecemb
  erlt/tdgtlt/trgt
• lt/tablegtlt/trgt