#knowledge-base-construction #machine-learning
A. Functional Requirements 1) Support for multiple types and formats of data. AKBC frameworks must offer the capability of process- ing a diversity of data and data formats. 2) Support for storage and search. The knowledge base framework must store extracted facts in a format that is indexable and queryable. 3) Support for flexible feature selection. To allow for variation and noise in input text, extraction rules should be flexible, and not rigid expressions or regex-like patterns. 4) Support for adding domain features. As there is variation between corpora from different domains, it must be possible to add domain-specific features to a knowledge base construction framework to increase the accuracy and completeness of a knowledge base. 5) Support for human feedback. For systems that require any user input, the knowledge base framework should support error analysis to fix (or flag) incorrect or overly- specific features.
If you want to change selection, open document below and click on "Move attachment"
pdf
cannot see any pdfsSummary
status | not read | | reprioritisations | |
---|
last reprioritisation on | | | suggested re-reading day | |
---|
started reading on | | | finished reading on | |
---|
Details