CSM06 Information Retrieval

1
CSM06 Information Retrieval

• LECTURE 4 Tuesday 19th October
• Dr Andrew Salway
• a.salway_at_surrey.ac.uk

2
LECTURE 4

• Part I Recap of Lecture 3 PageRank and other
  ranking factors review of Set Reading
• Part II Web IR continued
• Finding related pages by analysing link structure
  only
• Evaluating web search engines
• Some current RD Issues for web search engines
• Part III COURSEWORK

3
Finding Related Pages in the World Wide Web
(Dean and Henzinger 1999)

• Use a webpage (URL) as a query may be an easier
  way for a user to express their information need
• The user is saying I want more pages like this
  one maybe easier than thinking of good query
  words?
• e.g. the URL www.nytimes.com (New York Times
  newspaper) returns URLs for other newspapers and
  news organisations
• Aim is for high precision with fast execution
  using minimal information
• ?Two algorithms to find pages related to the
  query page using only connectivity information
  (nothing about webpage content or usage)
• Companion Algorithm
• Cocitation Algorithm

4
What does related mean?

• A related web page is one that addresses the
  same topic as the original page, but is not
  necessarily semantically identical

5
Companion Algorithm

• Based on Kleinbergs HITS algorithm mutually
  reinforcing authorities and hubs
• 1. Build a vicintiy graph for u
• 2. Contract duplicates and near-duplicates
• 3. Compute edge weights (i.e. links)
• 4. Compute hub and authority scores for each node
  (URL) in the graph ? return highest ranked
  authorities as results set

6
Companion Algorithm

• 1. Build a vicintiy graph for u
• The graph is made up of the following nodes and
  edges between them
• u
• Up to B parents of u, and for each parent up to
  BF of its children if u has gt B parents then
  choose randomly if a parent has gt BF children,
  then choose children closest to u
• Up to F children of u, and for each child up to
  FB of its parents
• NB. Use of a stop list of URLs with very high
  indegree

7
Companion Algorithm

• 2. Contract duplicates and near-duplicates if
  two nodes each have gt 10 links and gt 95 are in
  common then make them into one nodes whose links
  are the union of the two
• 3. Compute edge weights (i.e. links)
• Edges between nodes on the same host are weighted
  0
• Scaling to reduce the influence from any single
  host
• If there are k edges from documents on a first
  host to a single document on a second host then
  each edge has authority weight 1/k
• If there are l edges from a single document on a
  first host to a set of documents on a second
  host, we give each edge a hub weight of 1/l

8
Companion Algorithm

• 4. Compute hub and authority scores for each node
  (URL) in the graph ? return highest ranked
  authorities as results set
• a document that points to many others is a good
  hub, and a document that many documents point to
  is a good authority

9
Companion Algorithm

• 4. continued
• H hub vector with one element for the Hub value
  of each node
• A authority vector with one element for the
  Authority value of each node
• Initially all values set to 1

10
Companion Algorithm

• 4. continued
• Until H and A converge
• For all nodes n in the graph N
• An S Hnauthority_weight(n,n)
• For all nodes n in the graph N
• Hn S Anhub_weight(n,n)

11
Cocitation Algorithm

• Finds pages that are frequently cocited with the
  query web page u it finds other pages that are
  pointed to by many other pages that also point to
  u
• Two nodes are co-cited if they have a common
  parent the number of common parents is their
  degree of co-citation

12
Cocitation Algorithm

• Select up to B parents of u
• For each parent add up to BF of its children to
  the set of us siblings S
• Return nodes in S with highest degrees of
  cocitation with u
• NB. If lt 15 nodes in S that are cocited with u at
  least twice then restart using us URL with one
  path element removed, e.g. aaa.com/X/Y/Z ?
  aaa.com/X/Y

13
Evaluation

• 59 input URLs chosen by 18 volunteers (mainly
  computing professionals)
• The volunteers are shown results for each URL
  they chose and have to judge it 1 for valuable
  and 0 for not valuable
• ? Various calculations of precision, e.g.
  precision at 10 for the intersection group
  (those query URLs that all 3 algorithms returned
  results for)

14
Evaluation

• Authors suggest that their algorithms perform
  better than an algorithm (Netscapes) that
  incorporates content and usage information, as
  well as connectivity information This is
  surprising IS IT??
• Perhaps it is because they had more connectivity
  information??

15
Evaluation of Web Search Engines

• Precision may be applicable to evaluate a web
  search engine, but it may be the precision in the
  first page of results that is most important
• Recall, as traditionally defined, may not be
  applicable because it is difficult or impossible
  to identify all the relevant web-pages for a
  given query

16
Four strategies for evaluation of web search
engines

• Use precision and recall in the traditional way
  for a very tightly defined topic only applicable
  if all relevant web pages are known in advance
• Use relative recall estimate total number of
  relevant documents by doing a number of searches
  and adding the total number of relevant documents
  returned
• Statistically sample the web in order to estimate
  number of relevant pages
• Avoid recall altogether
• SEE Oppenheim, Morris and McKnight (2000), p. 194

17
Alternative Evaluation Criteria

• Number of web-pages covered, and coverage Is
  more pages covered better? May be more important
  that certain domains are included in coverage?
• Freshness / broken links Web-page content is
  frequently updated so index also needs to be
  updated broken links frustrate users. Should be
  relatively straightfoward to quantify.

18
Alternative Evaluation Criteria (continued)

• Search Syntax More experienced users may like
  the option of advanced searches, e.g. phrases,
  Boolean operators, and field searching.
• Human Factors and Interface Issues Evaluation
  from a users perspective is a more subjective
  criterion, however it is an important one it
  can be argued that an intuitive interface for
  formulating queries and interpreting results
  helps a user to get better results from the
  system.
• Quality of Abstracts related to interface issues
  are the abstracts of web-pages that a web
  search engine displays if good then these help
  a user to quickly identify more promising pages

19
Some current RD issues for web search engines

• Constant efforts to improve user interfaces, both
  to help users express their information needs and
  to help them understand more about the results,
  i.e. information visualisation More about this
  in Lecture 6
• Google Labs gives some insights to potentially up
  and coming features
• http//labs.google.com

20
Some things from Google Labs

• Desktop search
• search application that provides full text
  search over your email, computer files, chats,
  and the web pages you've viewed
• Since you can easily search information on your
  computer, you don't need to worry about
  organizing your files, email, or bookmarks

21
Some things from Google Labs

• SMS search
• Google SMS (Short Message Service) enables you
  to easily get precise answers to specialized
  queries from your mobile phone or device. Send
  your query as a text message and get phone book
  listings, dictionary definitions, product prices
  and more. Just text. No links. No web pages.

22
Some things from Google Labs

• Personalised Search
• Once youve entered a description of your general
  interests your search results are modified
  accordingly

23
(No Transcript)
24
Some things from Google Labs

• Google Sets
• Enter a few items and a longer list (of the same
  kinds of things) is returned

25
(No Transcript)
26
Set Reading

• Dean and Henzinger (1999), Finding Related Pages
  in the World Wide Web. Pages 1-10.
• http//citeseer.ist.psu.edu/dean99finding.html
• Oppenheim, Morris and McKnight (2000), The
  Evaluation of WWW Search Engines, Journal of
  Documentation, 56(2), pp. 190-211. SET READING
  is Pages 194-205. In Library Article Collection.

27
Exercise Googles Similar Pages

• It is suggested that Googles Similar Pages
  feature is based in part on the work of Dean and
  Henzinger. By making a variety of queries to
  Google and choosing Similar Pages see what you
  can find out about how this works.

28
Exercise web search engine evaluation

• Compare two or three web-search engines by making
  the same queries to each. How do they compare in
  terms of
• Coverage?
• Quality of highest ranked results?
• Ease of querying and understanding results?
• Ranking factors that they appear to be using?

29
Further Reading

• Rest of the papers given for Set Reading