Related Experiment Videos
Approaches to understanding the searching behavior of CrossFire users
Frank Cooke1, Nikolai Kopelev, Helen Schofield
1GlaxoSmithKline, King of Prussia, Pennsylvania 19406-0939, USA.
Journal of Chemical Information and Computer Sciences
|October 16, 2002
Summary
Analyzing user behavior in chemical information systems like CrossFire is key to improving search efficiency. A tool called CrossParse helps identify user difficulties and enhance training for better system utilization.
Area of Science:
- Information Science
- Cheminformatics
Background:
- Desktop chemical information systems are costly to purchase, support, and train users on.
- Understanding user search behavior is crucial for identifying and addressing inefficiencies.
Purpose of the Study:
- To analyze user search behavior within the CrossFire chemical information system.
- To identify deficiencies in search efficiency and effectiveness.
- To provide feedback for system enhancement and training improvement.
Main Methods:
- Development of a Log File Parser, CrossParse, in Visual Basic.
- Analysis of CrossFire log files to examine search activity, including structural feature building, search types, and result manipulation.
- Application of CrossParse at GlaxoSmithKline (GSK) and Manchester Information and Associated Services (MIMAS).
Main Results:
- CrossParse enables detailed analysis of user activity by individual, group, or population.
- Identified specific areas where users encounter difficulties with the CrossFire system.
- Facilitated comparison of search behavior between trained and untrained users.
Conclusions:
- CrossParse is effective in analyzing user search behavior in chemical information systems.
- The findings provide valuable feedback for CrossFire developers (MDL Inc.) to enhance the application.
- Insights gained can guide improvements in user training programs for chemical information systems.