Attention-Based Information Retrieval

1
Attention-Based Information Retrieval
Georg Buscher German Research Center for
Artificial Intelligence (DFKI) Knowledge
Management Department Kaiserslautern, Germany
SIGIR 07 Doctoral Consortium
2
Motivation
1
2
3

• Homer's personality is one of frequent stupidity,
  laziness, and explosive anger. He also suffers
  from a short attention span which complements his
  intense but short-lived passion for hobbies,
  enterprises and various causes. Furthermore, he
  is prone to emotional outbursts.

• Magnetic Resonance Imaging uses magnetic fields
  and radio waves to produce high quality two- or
  three-dimensional images of brain structures.
  Sensors read frequencies of radio waves and a
  computer uses the information to construct an
  image of the brain (see 2) .

• Positron Emission Tomography measures emissions
  from radioactively labeled metabolically active
  chemicals that have been injected into the
  bloodstream. The emission data are
  computer-processed to produce 2- or 3-dimensional
  images of the distribution of the chemicals
  throughout the brain. Especially useful are a
  wide array of chemicals used to map different
  aspects of neurotransmitter activity (see 3).

3
Outline

• Acquiring attention evidence
• Attention evidence through eye tracking
• Attention annotation and derivation with
  Dempster-Shafer
• Applications in Information Retrieval
• Attention-based TfIdf
• Context elicitation
• Context-based Index
• Query Expansion / result re-ranking

4
Sources of Attention-Data

• There are many indications of attention from the
  user

Reading evidence (implicit)
read
Annotations (explicit)
skimmed
longer viewed
5
Reading Detection An Example
6
Attention Annotations Imply Different Levels of
Attention

• Attention evidence values

0.7 1.0
0.5 1.0
1.0 1.0
0.2 0.7

• Range from 0 to 1
• Width of an interval expresses uncertainty

7
Dempster-Shafer Combination of Attention Evidence
The demo providedifferentvisualizationsan
d interfacesaccording situation. R R H R H
U R U R 0.5 1 0.85 1 0.96
1 0.85 1 0.5 1
Calculate one value of attention (att(t) bel(t)
0.2bel(t) 0.2pl(t))
0.6 0.88 0.97 0.88 0.6
In that way, the function att provides an
attention value for every term of the document.
attdifferent, d 0.88 attaccording, d
0.6 attsomethingElse, d 0
8
Outline

• Acquiring attention evidence
• Attention evidence through eye tracking
• Attention annotation and derivation with
  Dempster-Shafer
• Applications in Information Retrieval
• Attention-based TfIdf Desktop Index
• Context elicitation
• Context-based Index
• Query Expansion

9
Attention-Based Desktop Index

• A Desktop index is especially for re-finding
  known documents.
• You can better remember those parts of a document
  that you paid attention to.
• ? Attended terms should be weighted higher.
• TfIdf-based modification
• Attention is a local factor (like tf)
• The higher the maximal intensity of an attended
  document part, the more weight should be assigned
  to the attention value.
• The lower the maximal intensity of an attended
  document part, the more weight should be assigned
  to tf.

attention part
term frequency part
tft,d term frequency of term t in document d
a in 0 1 is a balancing factor for defining
the influence of attention in contrast to term
frequency.
attt,d attention value of term t in document d
10
Why Context? The Search for the Mental Model

• If a knowledge worker tries to recall something
  concerning a topic,does he primarily think
• on the basis of documents and document structures
  or
• on the basis of former thematic contexts?
• ? Rather the latter
• While re-finding some information, one does not
  search primarily for the document, but for the
  former mental model.Documents mediate.

11
Elicitation and Representation of the Thematic
Context
Document 1 Brain imaging
Document 2 Brain imaging
Document 3 The Simpsons

• Some read sub-documents
• Combination of the viewed sub-documents to one
  virtual context document (only those attended
  parts that have a thematic overlapping)

Document 4 Brain imaging
thematic context Brain imaging
12
Determination of Thematical Overlapping

• Determine buzzwords for each viewed document by
  using
• Attention value
• Idf of desktop index
• Compare buzzword vector with previous context
  vectors
• If there is a similarity, then merge with context
  vector
• Else buzzword vector is a new context

Currentlyvieweddocument(part)
?
Previouscontexts
13
Context-Based Vector-Space Index

• Common index structure

Doc1 Doc2 Doc3
Term1 Term2 Term3
0 1 0
4 0 1
2 3 1

• Idea two indexes1. Term Context 2. Context
  Document
• A context is represented by a virtual context
  document
• The value for each termcontext relation is
  influenced by the degree of attention

C1 C2 C3
Doc1 Doc2
Term1 Term2 Term3 Term4
5 2 0 1
2 1 0 3
1 2 1 3
C1 C2 C3
x x
x x
14
New Kinds of Search Tasks Possible

• Local searchFind for the current task (parts
  of) documents,that I formerly used for a similar
  task.
• Enterprise-wide searchFind for the current task
  (parts of) documents,that I do not know yet,
  butthat have been used by some colleague for a
  similar task.

15
Evaluation of the Context-Based Index

• Main advantage is expected to show up in several
  weeks.
• Not possible to do real-world eye tracking
  studies for such a long time
• Artificial experiment
• Several different exploration tasks within some
  hours
• Then some re-finding tasks about previously
  viewed content
• Measuring the time or user-satisfaction during
  the search process?

Context-based search
Normal search
16
Contextual Attention-Based Relevance Feedback

• Problem with context-based index it doesnt
  scale for web search? therefore query expansion
• Current elicited context (i.e. term vector)
  expresses current interest of the user
• Topmost characteristic keywords will be used for
  query expansion

17
The Global Picture
Eye Tracker
Attention data generation module
Attention-baseddesktop index
Text Mark Recognition
Attention-annotated document
Context-basedindex
Thank youfor your
Context document
attention
!
attention
Query expansionfor web search