Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
1
choose: we are not interested in product testing
assess is sometimes need to validate (political, economic decisions)
3
subpoint 1 is a corollary
subpoint two is a related claim
Polymer chemist wants chemical structures; medical researchers want graphs & tables
Fast moving research areas needs controlled vocabulary that tracks changes (e.g., renaming concepts; naming product lines, etc.)
5
NO User!
No Feedback!
6
8
The variability in the human and problem components is so great as to swamp any variance effects in other components
Hersh, W., Pentecost, J., & Hickam, D. (1996). A task-oriented approach to information retrieval evaluation, JASIS, 47, 50-56.
9
think aloud: Jorgensen & Liddy book indexes; Beaulieu OKAPI
transaction logs: BLS project, Marchionini & Hert
error analysis: Grateful Med project, Marchionini & Norman
time on task: # moves, time/screen, frequency of access (across sessions)--DO PEOPLE USE the SYSTEM?
cost-benefit: attention cycles
questionnaires: QUIS
simulations: Friedman, C. & Downs, S. Alternatives to current methods: Decision-theoretic approaches.
Jorgensen, C., & Liddy, E. (1996).
multiple retrieval efforts: see Voorhees data merging techniques
multiple evaluation efforts: see Perseus, BLS
10
accessibility: Florance dissertation--importance of tables, graphs; Dupont discussion (Rita Ayers) importance of chemical structures
interaction model: based on GOMS, see Dunlop SIGIR 97 paper
11
4