16 August 2017: Kobby Nuamah, 3rd Year Review Our objective is to expand the kinds of queries that can be asked and the range of answers that can be found in an automated query answering system. I'll explain: 1) our algorithm for RIF (Rich Inference Framework) and our internal representation of facts using frames, 2) the decomposition rules for expanding the search tree, and 3) the aggregate functions with which inferences (arithmetic and statistical) are drawn and propagated through the RIF tree to finally obtain an answer at the root. Additionally, I'll discuss our ideas for uncertainty in the QA process and explain how we account for uncertainty in our algorithm with the help of Bayesian statistics. Finally, I'll give a brief overview of our implementation of RIF, where our primary aim is to make it easy to extend RIF with new knowledge bases, decomposition rules and aggregate functions without having to modify the core RIF system. Aug 16 2017 14.00 - 15.00 16 August 2017: Kobby Nuamah, 3rd Year Review In this talk, I'll discuss our approach to query answering that goes beyond the traditional information retrieval techniques, to one where we make functional inferences using data retrieved at run-time from knowledge bases (both local and remote). IF 1.16
16 August 2017: Kobby Nuamah, 3rd Year Review Our objective is to expand the kinds of queries that can be asked and the range of answers that can be found in an automated query answering system. I'll explain: 1) our algorithm for RIF (Rich Inference Framework) and our internal representation of facts using frames, 2) the decomposition rules for expanding the search tree, and 3) the aggregate functions with which inferences (arithmetic and statistical) are drawn and propagated through the RIF tree to finally obtain an answer at the root. Additionally, I'll discuss our ideas for uncertainty in the QA process and explain how we account for uncertainty in our algorithm with the help of Bayesian statistics. Finally, I'll give a brief overview of our implementation of RIF, where our primary aim is to make it easy to extend RIF with new knowledge bases, decomposition rules and aggregate functions without having to modify the core RIF system. Aug 16 2017 14.00 - 15.00 16 August 2017: Kobby Nuamah, 3rd Year Review In this talk, I'll discuss our approach to query answering that goes beyond the traditional information retrieval techniques, to one where we make functional inferences using data retrieved at run-time from knowledge bases (both local and remote). IF 1.16
Aug 16 2017 14.00 - 15.00 16 August 2017: Kobby Nuamah, 3rd Year Review In this talk, I'll discuss our approach to query answering that goes beyond the traditional information retrieval techniques, to one where we make functional inferences using data retrieved at run-time from knowledge bases (both local and remote).
