CISA Seminar: Kobby Nuamah The increasing availability of knowledge bases (KBs) on the web has opened up the possibility for improved inference in automated query answering (QA) systems. We have developed a rich inference framework that responds to queries where no suitable answer is readily contained in any available data source. Most question answering and information retrieval systems assume that answers to queries are stored in some form in the KB, thereby limiting the range of answers they can find. We take an approach motivated by rich forms of inference using techniques, such as regression, for prediction. For instance, we can answer "what country in Europe is predicted to have the largest population in 2021?" by decomposing Europe geo-spatially, using regression on country population for past years and selecting the country with the largest predicted value. Our technique, which we refer to as Rich Inference, combines heuristics, logic and statistical methods to infer novel answers to queries. It also determines what facts are needed for inference, searches for them, and then integrates the diverse facts and their formalisms into the local query-specific inference tree. In this presentation, I will explain the internal representation of our framework, the grammar for expressing well-formed queries, the type signatures of inference methods that we use to compose a wide variety of queries, and the algorithm for the inference framework. I'll also show how we determine the confidence of answers inferred from this framework considering the various forms of uncertainty faced by the framework May 09 2016 13.30 - 15.00 CISA Seminar: Kobby Nuamah The Rich Inference Framework: Inferring Novel Answers in Query Answering. IF 4.31/4.33
CISA Seminar: Kobby Nuamah The increasing availability of knowledge bases (KBs) on the web has opened up the possibility for improved inference in automated query answering (QA) systems. We have developed a rich inference framework that responds to queries where no suitable answer is readily contained in any available data source. Most question answering and information retrieval systems assume that answers to queries are stored in some form in the KB, thereby limiting the range of answers they can find. We take an approach motivated by rich forms of inference using techniques, such as regression, for prediction. For instance, we can answer "what country in Europe is predicted to have the largest population in 2021?" by decomposing Europe geo-spatially, using regression on country population for past years and selecting the country with the largest predicted value. Our technique, which we refer to as Rich Inference, combines heuristics, logic and statistical methods to infer novel answers to queries. It also determines what facts are needed for inference, searches for them, and then integrates the diverse facts and their formalisms into the local query-specific inference tree. In this presentation, I will explain the internal representation of our framework, the grammar for expressing well-formed queries, the type signatures of inference methods that we use to compose a wide variety of queries, and the algorithm for the inference framework. I'll also show how we determine the confidence of answers inferred from this framework considering the various forms of uncertainty faced by the framework May 09 2016 13.30 - 15.00 CISA Seminar: Kobby Nuamah The Rich Inference Framework: Inferring Novel Answers in Query Answering. IF 4.31/4.33
May 09 2016 13.30 - 15.00 CISA Seminar: Kobby Nuamah The Rich Inference Framework: Inferring Novel Answers in Query Answering.
