ITEC 4020 M

1

• ITEC 4020 M
• Group 18
• Amna Al-Omari
• Divya Love
• Norbert Megler
• Omer Saleem
• Sachin Uppal
• Shahla Defileh

2
WEB SEARCH SYSTEM

• Presentation Overview
• Brief Overview of Assignment Objective
• Structure and Functionality
• Search Demonstration
• Questions

3
WEB SEARCH SYSTEM

• Introduction
• Our website, can be found at http//unix.aml.yorku
  .ca8080/w04_g18/search.jsp
• Our Web Search system is based on inverted file
  indexing using the XML document which has been
  created by the crawler that was supplied to us.
  Our site contains 3 main WebPages
• The main Search page which is built by JSP and
  contains a text box and 2 buttons (reset and
  submit).
• The result page which is built by JSP and
  contains all the hyperlinks for all documents
  that hold the keyword.
• The display page which is built by Xml and
  displays the clicked on document.

4
WEB SEARCH SYSTEM

• Logical structure
• 1- java class which will read the given XML file
  and split it into 1139 separate XML documents.
• - We read the XML file using FileInputStream and
  BufferedReader.
• - The file is read one line at a time and each
  line is compared to the index ltPubmedArticlegt
  which signals the beginning of a new article.
• - Upon detection of word a new XML document file
  is created
• - The file number is kept track off and once the
  whole article is written into a file, the file
  counter is incremented by one.

5
WEB SEARCH SYSTEM

• CONTINUED
• 2- create a Temporary (merged) file which goes
  through the entire 1100 document and identifies
• all terms
• their document number
• frequency

6
WEB SEARCH SYSTEM

• Continued..
• 3- Next we create the First level index.
• - It uses a simple java class which reads from
  the Temporary file.
• - This index includes a counter (which
  represents the total number of terms), the terms
  which appear only once, number of the document
  that includes that specific term, and the total
  number of frequency of each term all of this is
  then written into a text file.

7

• Continued
• 4- Next is the creation of the second level
  indexing, created by a simple java class which
  includes counter, term, document number, and
  frequency.

8

• Searching Functionality
• We are using MVC (model view controller)
  architecture i.e. Servlet acting as controller,
  JSP is used for displaying results and Java Bean
  has the main business logic. Once the user
  submits the keyword, the search functionality
  goes through the first level index to find the
  counter number for that specific term and then
  matches that counter number with all the XML
  document number which appears in the second
  index. Then goes through all the XML documents
  and grabs all the relevant document title for
  display.

9

• Displaying of Results.
• XSL files take care of this functionality. Once
  the user clicks on any hyperlink that specific
  document displays through XML which uses XSL file.

10

• The End
• Questions?
• Thank You