Main Page > Knowledge Extractor > Performance of version 0.1

Ideally, we want to create a quantitative performance function to measure the accuracy and quality of our knowledge base. However, because this task could be a separate project in itself, we have decided on a qualitative means of evaluating our knowledge base for the time being.

We have chosen 11 words to use as performance measures and will compare queries of these words between different versions of the knowledge base. Below are the 11 words and the results of querying them against version 0.1 of our knowledge base.

Words and Results - version 0.1

food

yoshi

whiteboard

football

University of Illinois

car

China

cellphone

president

bookbag

twins

Comments

Please comment below about the quality and precision of the queries. Talk about what went right and what went wrong.