John Prager

Major advances in Question Answering technology were needed for IBM Watson to play Jeopardy! at championship level -- the show requires rapid-fire answers to challenging natural language questions, broad general knowledge, high precision, and accurate confidence estimates. In addition, Jeopardy! features four types of decision making carrying great strategic importance: (1) Daily Double wagering; (2) Final Jeopardy wagering; (3) selecting the next square when in control of the board; (4) deciding whether to attempt to answer, i.e., "buzz in." Using sophisticated strategies for these decisions, that properly account for the game state and future event probabilities, can significantly boost a players overall chances to win, when compared with simple "rule of thumb" strategies. This article presents our approach to developing Watsons game-playing strategies, comprising development of a faithful simulation model, and then using learning and Monte-Carlo methods within the simulator to optimize Watsons strategic decision-making. After giving a detailed description of each of our game-strategy algorithms, we then focus in particular on validating the accuracy of the simulators predictions, and documenting performance improvements using our methods. Quantitative performance benefits are shown with respect to both simple heuristic strategies, and actual human contestant performance in historical episodes. We further extend our analysis of human play to derive a number of valuable and counterintuitive examples illustrating how human contestants may improve their performance on the show.

Although the majority of evidence analysis in DeepQA is focused on unstructured information (e.g., natural-language documents), several components in the DeepQA system use structured data (e.g., databases, knowledge bases, and ontologies) to generate potential candidate answers or find additional evidence. Structured data analytics are a natural complement to unstructured methods in that they typically cover a narrower range of questions but are more precise within that range. Moreover, structured data that has formal semantics is amenable to logical reasoning techniques that can be used to provide implicit evidence. The DeepQA system does not contain a single monolithic structured data module; instead, it allows for different components to use and integrate structured and semistructured data, with varying degrees of expressivity and formal specificity. This paper is a survey of DeepQA components that use structured data. Areas in which evidence from structured sources has the most impact include typing of answers, application of geospatial and temporal constraints, and the use of formally encoded a priori knowledge of commonly appearing entity types such as countries and U.S. presidents. We present details of appropriate components and demonstrate their end-to-end impact on the IBM Watsoni system.

The top-performing Question–Answering (QA) systems have been of two types: consistent, solid, well-established and multi-faceted systems that do well year after year, and ones that come out of nowhere employing totally innovative approaches and which out-perform almost everybody else. This article examines both types of system in depth. We establish what a “typical” QA-system looks like, and cover the commonly used approaches by the component modules. Understanding this will enable any proficient system developer to build his own QA-system. Fortunately there are many components available for free from their developers to make this a reasonable expectation for a graduate-level project. We also look at particular systems that have performed well and which employ interesting and innovative approaches.

We describe a hybrid approach to improving search performance by providing a natural language front end to a traditional keyword-based search engine. The key component of the system is iterative query formulation and retrieval, in which one or more queries are automatically formulated from the user's question, issued to the search engine, and the results accumulated to form the hit list. New queries are generated by relaxing previously-issued queries using transformation rules, applied in an order obtained by reinforcement learning. This statistical component is augmented by a knowledge-driven hub-page identifier that retrieves a hub-page for the most salient noun phrase in the question, if possible. Evaluation on an unseen test set over the www.ibm.com public website with 1.3 million webpages shows that both components make substantial contribution to improving search performance, achieving a combined 137% relative improvement in the number of questions correctly answered, compared to a baseline of keyword queries consisting of two noun phrases.

In this paper, we describe a system to rank suspected answers to natural language questions. We process both corpus and query using a new technique, predictive annotation, which augments phrases in texts with labels anticipating their being targets of certain kinds of questions. Given a natural language question, an IR system returns a set of matching passages, which are then analyzed and ranked according to various criteria described in this paper. We provide an evaluation of the techniques based on results from the TREC Q&A evaluation in which our system participated.