Hierarchical task topology for retrieving information from within a simulated information ecosystem
Journal of Network and Computer Applications - Special issue on computational intelligence on the internet
| Hi-index | 0.00 |
The design of the Tocorime Apicu integrated search engine draws upon the honeybee information sharing model, which is an inherently parallel, hierarchical model for task distribution and communication. The hierarchical structure is the foundation to the clustered approach for delegating and partitioning tasks. These tasks include foraging the Internet for information, assessing the computational effort required to retrieve information in the form of Web pages, sharing information among indexer nodes, and extracting knowledge from the information. A variety of search engines and their respective information retrieval systems are compared with the information sharing model to show the usefulness of this latter approach. The advent of WWW search engines addressed the shortcomings of early information retrieval systems. However, current systems are hampered by their reliance on outdated information, the result of an Internet retrieval approach for gathering new and updated pages as opposed to a resource discovery system based on the information sharing model. The honeybee information sharing model provides a solution to these problems, and provides an alternative model for what is really needed to index the WWW. This model uses hierarchical communication mechanisms to convey resource requirements associated with tasks needed for efficient use of information. Unique aspects of the honeybee model have been adapted as building blocks for the Tocorime Apicu search engine. The word Tocorime, meaning “spirit,” comes from an ancient Amazon Indian language. Apicu comes from the Latin apis cultura, meaning “honeybee culture” or “the study of honeybees.” The phrase Tocorime Apicu is used as “in the spirit of bee culture.”