Watsonpaths: Scenario-Based Question Answering and Inference Over Unstructured Information

Total Page:16

File Type:pdf, Size:1020Kb

Watsonpaths: Scenario-Based Question Answering and Inference Over Unstructured Information Articles WatsonPaths: Scenario-Based Question Answering and Inference over Unstructured Information Adam Lally, Sugato Bagchi, Michael A. Barborak, David W. Buchanan, Jennifer Chu-Carroll, David A. Ferrucci, Michael R. Glass, Aditya Kalyanpur, Erik T. Mueller, J. William Murdock, Siddharth Patwardhan, John M. Prager n We present WatsonPaths, a novel BM Watson is a question-answering system that takes nat - system that can answer scenario-based ural language questions as input and produces precise questions. These include medical ques - Ianswers along with accurate confidences as output (Fer - tions that present a patient summary rucci et al. 2010). In 2011, in a modified version of the quiz and ask for the most likely diagnosis or show Jeopardy!, Watson defeated two of the best human play - most appropriate treatment. Watson - Paths builds on the IBM Watson ques - ers. tion-answering system. WatsonPaths Jeopardy! questions are usually factoid questions — the breaks down the input scenario into answer and supporting evidence are usually stated explicitly individual pieces of information, asks in some document in the corpus. While in practice we may relevant subquestions of Watson to con - retrieve multiple redundant documents, in principle the clude new information, and represents answer could be expressed succinctly in one. The main chal - these results in a graphic model. Proba - lenges for a factoid question-answering system are retrieving bilistic inference is performed over the the correct document, and then extracting the correct answer graph to conclude the answer. On a set of medical test preparation questions, from the document. At the core of Watson’s question answer - WatsonPaths shows a significant ing is a suite of algorithms that match passages containing improvement in accuracy over multiple candidate answers to the original question. These algorithms baselines. have been described in a series of articles (Chu-Carroll et al. 2012; Ferrucci 2012; Gondek et al. 2012; Lally et al. 2012; McCord, Murdock, and Boguraev 2012; Murdock et al. 2012a; 2012b). In some important applications, however, questions do not have this “factoid” character. Consider the Copyright © 2017, Association for the Advancement of Artificial Intelligence. All rights reserved. ISSN 0738-4602 Summer 2017 59 Articles “A 32-year-old woman with type 1 diabetes mellitus has had progressive renal failure... Her hemoglobin concentration is 9 g/dL... A blood smear shows normochromic, normocytic cells. What is the problem? Patient’s hemoglobin conc. Patient’s blood smear Patient has renal failure is 9 g/dL [low] shows normocytic cells Evidence: “Low hemoglobin conc. indicates anemia.” Evidence: “Erythropoietin is produced in the kidneys.” Patient has anemia Evidence: “Normocytic anemia is a type of anemia Patient is at risk for with normal red blood Erythropoietin de!ciency cells.” Patient has normocytic anemia Evidence: “Erythropoietin de!ciency is a cause of normocytic anemia.” Most likely cause of low hemoglobin conc. is Erythropoietin de!ciency Figure 1. A Simple Diagnosis Graph for a Patient with Erythropoietin Deficiency. following questions, one from medicine and one knowledge to a specific case, as in a medical scenario from taxation: about a patient. A 32-year-old woman with type 1 diabetes mellitus has Before beginning work on automated scenario- had progressive renal failure. Her hemoglobin con - based question answering, we investigated how centration is 9 g/dL. A blood smear shows nor - humans solve such questions. We asked domain mochromic, normocytic cells. What is the problem? experts to describe their approach to solving a set of I inherited real-estate from a relative who died 5 years scenario-based questions in the medical domain. An ago via a trust that was created before his death. The example is shown in figure 1. Many drew a graph of property was sold this year after dissolution of the initial signs and symptoms leading to their most like - trust, and the money was put in a Roth-IRA. Which ly possible causes and connecting them to a final tax form(s) do I need to file? conclusion. This motivated us to look into graph- We will call these types of questions scenario-based based methods as a way of answering scenario-based questions. In these types of questions, it is not gener - questions automatically. ally the case that the answer and supporting evidence In this article, we describe WatsonPaths, a system can be contained in one document. Rather, for many that builds on Watson to answer scenario-based ques - scenario-based questions, information from multiple tions. The core idea is to break the question down documents and other sources must generally be into parts, over which we can ask and answer factoid retrieved and then integrated to answer the questions subquestions using Watson, then integrate these properly. Furthermore, we must often apply general answers into a graphic model that can be used to 60 AI MAGAZINE Articles answer the larger scenario-based question. We show Scenario Analysis that WatsonPaths not only outperforms a baseline The first step in the pipeline is scenario analysis, system that uses simple information retrieval, but where we identify factors in the input scenario that also outperforms its own subcomponent, Watson, in may be of importance. In the medical domain, the answering a set of scenario-based questions from the factors may include demographics (“32-year old medical domain. woman”), preexisting conditions (“type 1 diabetes mellitus”), signs and symptoms (“progressive renal WatsonPaths Medical Use Case failure”), and test results (“hemoglobin concentra - tion is 9 g/dL,” “normochromic cells,” “normocytic Although WatsonPaths is intended as a domain-gen - cells”). The extracted factors become nodes in a eral technology for scenario-based question answer - graph structure called the assertion graph, on which ing, we decided to start by focusing our attention on the remaining steps of the process will operate. the medical domain. We focused on the problem of patient scenario analysis, where the goal is typically Node Prioritization a diagnosis or a treatment recommendation. The next step is node prioritization, where we decide To explore this kind of problem solving, we which nodes in the graph are most important for obtained a set of medical test preparation questions. solving the problem. In a small scenario like this These are multiple-choice medical questions based example, we may be able to explore everything, but on an unstructured or semistructured natural lan - in general this will not be the case. Factors that affect guage description of a patient. Although Watson - the priority of a node may include the system’s con - Paths is not restricted to multiple-choice questions, fidence in the node assertion or the system’s estima - we saw multiple-choice questions as a good starting tion of how fruitful it would be to expand a node. For point for development. Many of these questions example, normal test results and demographic infor - involve diagnosis, either as the entire question, as in mation are generally less useful for starting a diagno - the previous medical example, or as an intermediate sis than symptoms and abnormal test results. step, as in the following example: A 63-year old patient is sent to the neurologist with a Relation Generation clinical picture of resting tremor that began 2 years The relation-generation step builds the assertion ago. At first it was only on the left hand, but now it graph. We do this primarily by asking Watson ques - compromises the whole arm. At physical exam, the tions about the factors. In medicine we want to know patient has an unexpressive face and difficulty in the causes of the findings and abnormal test results walking, and a continuous movement of the tip of the that are consistent with the patient’s demographic first digit over the tip of the second digit of the left information and normal test results. Given the sce - hand is seen at rest. What part of his nervous system nario in the Introduction, we could ask, “What does is most likely affected? type 1 diabetes mellitus cause?” We use a medical For this question, it is useful to diagnose that the ontology to guide the process of formulating sub - patient has Parkinson’s disease before determining questions to ask Watson. Relevant factors may also be which part of his nervous system is most likely affect - combined to form a single, more targeted question. ed. These multistep inferences are a natural fit for the Because in this step we want to emphasize recall, we graphs that WatsonPaths constructs. In this example, take several of Watson’s highly ranked answers. The the diagnosis is the missing link on the way to the exact number of answers taken, or the confidence final answer. threshold, are parameters that must be tuned. Given a set of answers, we add them to the graph as nodes, Scenario-Based Question Answering with edges from nodes that were used in questions to nodes that were answers. The edge is labeled with the In scenario-based question answering, the system predicate used to formulate the question (like causes receives a scenario description that ends with a or indicates ), and the strength of the edge is initially punch line question. For instance, the punch line set to Watson’s confidence in the answer. question in the Parkinson’s example is “What part of Although Watson is the primary way we add edges his nervous system is most likely affected?” Instead of to the graph, WatsonPaths allows for any number of treating the entire scenario as one monolithic ques - relation generator components to post edges to the tion as would Watson, WatsonPaths explores multi - graph. For instance, we apply term matchers to pairs ple facts in the scenario in parallel and reasons with of nodes, and post a relation between nodes that the results of its exploration as a whole to arrive at match.
Recommended publications
  • Open Mind Common Sense: Knowledge Acquisition from the General Public
    Open Mind Common Sense: Knowledge Acquisition from the General Public Push Singh, Grace Lim, Thomas Lin, Erik T. Mueller Travell Perkins, Mark Tompkins, Wan Li Zhu MIT Media Laboratory 20 Ames Street Cambridge, MA 02139 USA {push, glim, tlin, markt, wlz}@mit.edu, [email protected], [email protected] Abstract underpinnings for commonsense reasoning (Shanahan Open Mind Common Sense is a knowledge acquisition 1997), there has been far less work on finding ways to system designed to acquire commonsense knowledge from accumulate the knowledge to do so in practice. The most the general public over the web. We describe and evaluate well-known attempt has been the Cyc project (Lenat 1995) our first fielded system, which enabled the construction of which contains 1.5 million assertions built over 15 years a 400,000 assertion commonsense knowledge base. We at the cost of several tens of millions of dollars. then discuss how our second-generation system addresses Knowledge bases this large require a tremendous effort to weaknesses discovered in the first. The new system engineer. With the exception of Cyc, this problem of scale acquires facts, descriptions, and stories by allowing has made efforts to study and build commonsense participants to construct and fill in natural language knowledge bases nearly non-existent within the artificial templates. It employs word-sense disambiguation and intelligence community. methods of clarifying entered knowledge, analogical inference to provide feedback, and allows participants to validate knowledge and in turn each other. Turning to the general public 1 In this paper we explore a possible solution to this Introduction problem of scale, based on one critical observation: Every We would like to build software agents that can engage in ordinary person has common sense of the kind we want to commonsense reasoning about ordinary human affairs.
    [Show full text]
  • Automatic Affective Feedback in an Email Browser
    Automatic Affective Feedback in an Email Browser Hugo Liu Henry Lieberman Ted Selker Software Agents Group Software Agents Group Context-Aware Computing Group MIT Media Laboratory MIT Media Laboratory MIT Media Laboratory Cambridge, MA 02139 Cambridge, MA 02139 Cambridge, MA 02139 +1 617 253 5334 +1 617 253 0315 +1 617 253 6968 [email protected] [email protected] [email protected] ABSTRACT delighted us, the text sits unmoved in cold, square boxes on This paper demonstrates a new approach to recognizing and the computer screen. Nass et al.’s study of human- presenting the affect of text. The approach starts with a computer social interaction reveals that people naturally corpus of 400,000 responses to questions about everyday expect their interactions with computers to be social and life in Open Mind Common Sense. This so-called affective, just as with other people! [20],[21]. commonsense knowledge is the basis of a textual affect Sadly though, people have been so conditioned to expect so sensing engine. The engine dynamically analyzes a user’s little from the user interfaces of today that we are not even text and senses broad affective qualities of the story at the bothered by their inability to affectively respond to us like a sentence level. This paper shows how a commonsense friend or family member might do. affect model was constructed and incorporated into Chernov face style feedback in an affectively responsive This shortcoming in current user interfaces hinders email browser called EmpathyBuddy. This experimental progress in the bigger picture too. If software is to system reacts to sentences as they are typed.
    [Show full text]
  • Bootstrapping Commonsense Knowledge
    ASTRID: Bootstrapping Commonsense Knowledge Hans Peter Willems MIND|CONSTRUCT March 2021 Abstract The need for Commonsense Knowledge in the machine becomes more and more apparent, as we try to move forward in the development of Artificial (General) Intelligence. It is becoming evident that this kind of knowledge is paramount in the human cognitive capacity and therefore also crucial for machine intelligence to ever reach any level of performance nearing that of a human brain. However, attaining a sufficient amount, and qualitative level, of human Commonsense Knowledge in the machine, appears to be an ‘AI­hard’ or ‘AI­complete’ problem. How do humans do this? There is a lot to be learned from child development, and although there are AI­projects that try to use a developmental model to ‘grow’ intelligence, there have not been any (relevant) Commonsense Knowledge projects that leveraged the child development paradigm. That is, until now. I present ASTRID (Analysis of Systemic Tagging Results in Intelligent Dynamics), a real­world implementation of a Cognitive Architecture based on human developmental models. The current state of this project is the result of a full decade of research and development. This paper describes the project background, underlying philosophies, objectives and current results and insights. Keywords: Commonsense Knowledge, Unsupervised Transfer Learning, Machine Intelligence, Semantics, Natural Language Processing, Deep Inference. ©2021 Hans Peter Willems ­ First published March 22, 2021 online @ https://www.mindconstruct.com This work is licensed under a Creative Commons Attribution­ShareAlike 4.0 License 2 ASTRID: Bootstrapping Commonsense Knowledge The case for Commonsense Knowledge As early as 1959, John McCarthy argued for the need of Commonsense Knowledge to attain human level Artificial Intelligence (McCarthy, 1959), currently referred to as Artificial General Intelligence (AGI).
    [Show full text]
  • Natural Language Understanding with Commonsense Reasoning
    E.T.S. DE INGENIEROS INFORMÁTICOS UNIVERSIDAD POLITÉCNICA DE MADRID MASTER TESIS MSc IN ARTIFICIAL INTELLIGENCE (MUIA) NATURAL LANGUAGE UNDERSTANDING WITH COMMONSENSE REASONING: APPLICATION TO THE WINOGRAD SCHEMA CHALLENGE AUTHOR: ALFONSO LÓPEZ TORRES SUPERVISOR: MARTÍN MOLINA GONZÁLEZ JUNE, 2016 This is for my children Carla and Alonso, and my wife Véronique Thanks for their unconditional support and patient (also for the coming adventures…) v Acknowledgments: I’d like to thank the advices and help received from Martín. I was very lucky being your student. vi RESUMEN En 1950, Alan Turing propuso un test para evaluar el grado de inteligencia humana que podría presentar una máquina. La idea principal era realmente sencilla: llevar a cabo una charla abierta entre un evaluador y la máquina. Si dicho evaluador era incapaz de discernir si el examinado era una persona o una máquina, podría afirmarse que el test había sido superado. Desde entonces, a lo largo de los últimos 60 años se han presentado numerosas propuestas a través de los cuales se han puesto al descubierto ciertas debilidades del test. Quizás la más importante es el hecho de centrarse en la inteligencia humana, dejando a un lado otros tipos de inteligencia. El test obliga en gran medida a definir en la máquina un comportamiento antropomórfico y de imitación con el único fin de pasar el test. Con el fin de superar estos y otros puntos débiles, Hector Levesque propuso en 2011 un nuevo reto, “The Winograd Schema Challenge”. Un sencillo test basado en Pregunta y Respuesta sobre una frase que describe una situación cotidiana.
    [Show full text]
  • Federated Ontology Search Vasco Calais Pedro CMU-LTI-09-010
    Federated Ontology Search Vasco Calais Pedro CMU-LTI-09-010 Language Technologies Institute School of Computer Science Carnegie Mellon University 5000 Forbes Ave. Pittsburgh, PA 15213 www.lti.cs.cmu.edu Thesis Committee: Jaime Carbonell, Chair Eric Nyberg Robert Frederking Eduard Hovy, Information Sciences Institute Submitted in partial fulfillment of the requirements for the degree Doctor of Philosophy In Language and Information Technologies Copyright © 2009 Vasco Calais Pedro For my grandmother, Avó Helena, I am sorry I wasn’t there Abstract An Ontology can be defined as a formal representation of a set of concepts within a domain and the relationships between those concepts. The development of the semantic web initiative is rapidly increasing the number of publicly available ontologies. In such a distributed environment, complex applications often need to handle multiple ontologies in order to provide adequate domain coverage. Surprisingly, there is a lack of adequate frameworks for enabling the use of multiple ontologies transparently while abstracting the particular ontological structures used by that framework. Given that any ontology represents the views of its author or authors, using multiple ontologies requires us to deal with several significant challenges, some stemming from the nature of knowledge itself, such as cases of polysemy or homography, and some stemming from the structures that we choose to represent such knowledge with. The focus of this thesis is to explore a set of techniques that will allow us to overcome some of the challenges found when using multiple ontologies, thus making progress in the creation of a functional information access platform for structured sources.
    [Show full text]
  • Probabilistic Approaches for Answer Selection in Multilingual Question Answering
    Probabilistic Approaches for Answer Selection in Multilingual Question Answering Jeongwoo Ko Aug 27 2007 Language Technologies Institute School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Thesis Committee: Eric Nyberg (Chair) Teruko Mitamura Jaime Carbonell Luo Si (Purdue University) Copyright c 2007 Jeongwoo Ko This work was supported in part by ARDA/DTO Advanced Question Answering for Intelligence (AQUAINT) program award number NBCHC040164 and used the NTCIR 5-6 corpus. Keywords: Answer ranking, answer selection, probabilistic framework, graphi- cal model, multilingual question answering To my family for love and support. iv Abstract Question answering (QA) aims at finding exact answers to a user’s natural language question from a large collection of documents. Most QA systems combine information retrieval with extraction techniques to iden- tify a set of likely candidates and then utilize some selection strategy to generate the final answers. This selection process can be very challenging, as it often entails ranking the relevant answers to the top positions. To address this challenge, many QA systems have incorporated semantic re- sources for answer ranking in a single language. However, there has been little research on a generalized probabilistic framework that models the correctness and correlation of answer candidates for multiple languages. In this thesis, we propose two probabilistic models for answer ranking: independent prediction and joint prediction. The independent prediction model directly estimates the probability of an individual answer candi- date given the degree of answer relevance and the amount of supporting evidence provided in a set of answer candidates. The joint prediction model uses an undirected graph to estimate the joint probability of all answer candidates, from which the probability of an individual candidate is inferred.
    [Show full text]
  • Improving User Experience in Information Retrieval Using Semantic Web and Other Technologies Erfan Najmi Wayne State University
    Wayne State University Wayne State University Dissertations 1-1-2016 Improving User Experience In Information Retrieval Using Semantic Web And Other Technologies Erfan Najmi Wayne State University, Follow this and additional works at: https://digitalcommons.wayne.edu/oa_dissertations Part of the Computer Sciences Commons Recommended Citation Najmi, Erfan, "Improving User Experience In Information Retrieval Using Semantic Web And Other Technologies" (2016). Wayne State University Dissertations. 1654. https://digitalcommons.wayne.edu/oa_dissertations/1654 This Open Access Dissertation is brought to you for free and open access by DigitalCommons@WayneState. It has been accepted for inclusion in Wayne State University Dissertations by an authorized administrator of DigitalCommons@WayneState. IMPROVING USER EXPERIENCE IN INFORMATION RETRIEVAL USING SEMANTIC WEB AND OTHER TECHNOLOGIES by ERFAN NAJMI DISSERTATION Submitted to the Graduate School of Wayne State University, Detroit, Michigan in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY 2016 MAJOR: COMPUTER SCIENCE Approved By: Advisor Date ⃝c COPYRIGHT BY ERFAN NAJMI 2016 All Rights Reserved ACKNOWLEDGEMENTS I would like to express my heartfelt gratitude to my PhD advisor, Dr. Zaki Malik, for supporting me during these past years. I could not have asked for a better advisor and a friend, one that let me choose my path, help me along it and has always been there if I needed to talk to a friend. I really appreciate all the time he spent and all the patience he showed towards me. Secondly I would like to thank my committee members Dr. Fengwei Zhang, Dr. Alexander Kotov and Dr. Abdelmounaam Rezgui for the constructive feedback and help they provided.
    [Show full text]
  • Efficient Hierarchical Entity Classifier Using Conditional Random Fields
    Efficient Hierarchical Entity Classifier Using Conditional Random Fields Koen Deschacht Marie-Francine Moens Interdisciplinary Centre for Law & IT Interdisciplinary Centre for Law & IT Katholieke Universiteit Leuven Katholieke Universiteit Leuven Tiensestraat 41, 3000 Leuven, Belgium Tiensestraat 41, 3000 Leuven, Belgium [email protected] [email protected] Abstract Rigau, 1996; Yarowsky, 1995), where the sense for a word is chosen from a much larger inventory In this paper we develop an automatic of word senses. classifier for a very large set of labels, the We will employ a probabilistic model that’s WordNet synsets. We employ Conditional been used successfully in NER (Conditional Ran- Random Fields (CRFs) because of their dom Fields) and use this with an extensive inven- flexibility to include a wide variety of non- tory of word senses (the WordNet lexical database) independent features. Training CRFs on a to perform entity detection. big number of labels proved a problem be- In section 2 we describe WordNet and it’s use cause of the large training cost. By tak- for entity categorization. Section 3 gives an ing into account the hypernym/hyponym overview of Conditional Random Fields and sec- relation between synsets in WordNet, we tion 4 explains how the parameters of this model reduced the complexity of training from are estimated during training. We will drastically 2 2 O(T M NG) to O(T (logM) NG) with reduce the computational complexity of training in only a limited loss in accuracy. section 5. Section 6 describes the implementation 1 Introduction of this method, section 7 the obtained results and finally section 8 future work.
    [Show full text]
  • Simulating Human Associations with Linked Data – End-To- End Learning of Graph Patterns with an Evolutionary Algorithm Supervisors: Prof
    dissertation SIMULATINGHUMANASSOCIATIONSWITH LINKEDDATA End-to-End Learning of Graph Patterns with an Evolutionary Algorithm sdog tcat ni ... nj Thesis approved by the Department of Computer Science of the TU Kaiserslautern for the award of the Doctoral Degree doctor of natural sciences (dr. rer. nat.) to Jörn Hees Date of the viva: 2018-04-09 Dean: Prof. Dr. Stefan Deßloch Reviewers: Prof. Dr. Prof. h.c. Andreas Dengel Prof. Dr. Heiko Paulheim (University of Mannheim) D 386 Jörn Hees: Simulating Human Associations with Linked Data – End-to- End Learning of Graph Patterns with an Evolutionary Algorithm supervisors: Prof. Dr. Prof. h.c. Andreas Dengel Prof. Dr. Heiko Paulheim (University of Mannheim) supplemental material: https://w3id.org/associations or http://purl.org/associations contact information: http://joernhees.de ABSTRACT In recent years, enormous progress has been made in the field of Ar- tificial Intelligence (AI). Especially the introduction of Deep Learning and end-to-end learning, the availability of large datasets and the nec- essary computational power in form of specialised hardware allowed researchers to build systems with previously unseen performance in areas such as computer vision, machine translation and machine gam- ing. In parallel, the Semantic Web and its Linked Data movement have published many interlinked RDF datasets, forming the world’s largest, decentralised and publicly available knowledge base. Despite these scientific successes, all current systems are still nar- row AI systems. Each of them is specialised to a specific task and cannot easily be adapted to all other human intelligence tasks, as would be necessary for Artificial General Intelligence (AGI).
    [Show full text]
  • Redalyc.Cognitive Modules of an NLP Knowledge Base for Language
    Procesamiento del Lenguaje Natural ISSN: 1135-5948 [email protected] Sociedad Española para el Procesamiento del Lenguaje Natural España Periñán-Pascual, Carlos; Arcas-Túnez, Francisco Cognitive Modules of an NLP Knowledge Base for Language Understanding Procesamiento del Lenguaje Natural, núm. 39, 2007, pp. 197-204 Sociedad Española para el Procesamiento del Lenguaje Natural Jaén, España Available in: http://www.redalyc.org/articulo.oa?id=515751739024 How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Procesamiento del Lenguaje Natural, nº39 (2007), pp. 197-204 recibido 30-04-2007; aceptado 22-06-2007 Cognitive Modules of an NLP Knowledge Base for Language Understanding Carlos Periñán-Pascual Francisco Arcas-Túnez Universidad Católica San Antonio Universidad Católica San Antonio Campus de los Jerónimos s/n Campus de los Jerónimos s/n 30107 Guadalupe - Murcia (Spain) 30107 Guadalupe - Murcia (Spain) [email protected] [email protected] Resumen : Algunas aplicaciones del procesamiento del lenguaje natural, p.ej. la traducción automática, requieren una base de conocimiento provista de representaciones conceptuales que puedan reflejar la estructura del sistema cognitivo del ser humano. En cambio, tareas como la indización automática o la extracción de información pueden ser realizadas con una semántica superficial. De todos modos, la construcción de una base de conocimiento robusta garantiza su reutilización en la mayoría de las tareas del procesamiento del lenguaje natural. El propósito de este artículo es describir los principales módulos cognitivos de FunGramKB, una base de conocimiento léxico-conceptual multipropósito para su implementación en sistemas del procesamiento del lenguaje natural.
    [Show full text]
  • Using and Interfacing Background Knowledge in Story Understanding
    Using and Interfacing Background Knowledge in Story Understanding Nemecio R. Chavez, Jr.1, Heather D. Pfeiffer2, Roger T. Hartley1 1Department of Computer Science, 2Klipsch School of Electrical and Computer Engineering, New Mexico State University Box 30001, MSC CS/3-O, Las Cruces, NM, 88003-8001 USA {nchavez,hdp,rth}@cs.nmsu.edu Abstract. This paper details the use of background knowledge within a story understanding system. The story understanding system is based upon a multi-agent system (MAS). The MAS combines different forms of knowledge into a meaningful structure from which understanding can be demonstrated, i.e., question answering. The system uses two forms of knowledge to accomplish this task: 1) knowledge about objects in the world; and 2) prototypes. The latter represents higher-order knowledge or experience such as repeating a process and forms of thinking like abduction and deduction. The former stores knowledge about objects in the real world and their relationship to each other. The two forms are viewed as two distinct forms of data that might be used in our own understanding process, thus, they are treated separately by the system. Keywords Knowledge, knowledge bases, databases, communication, story understanding. 1 INTRODUCTION Described are two forms of knowledge used to help understand a simple children’s story. The bottom layer of knowledge provides basic knowledge needed for story understanding. The basic knowledge is information about the world that a 6 or 7 year old child might know and use. The story under- standing system also provides an added layer of knowledge that will be referred to as prototypes.
    [Show full text]
  • Automatic Conversion of Natural Language to 3D Animation
    Automatic Conversion of Natural Language to 3D Animation Minhua Ma B.A., M.A., M.Sc. Faculty of Engineering University of Ulster A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy July 2006 ii Table of Contents List of Figures vi List of Tables ix Acknowledgements x Abstract xi Abbreviations xii Note on access to contents xiv 1. INTRODUCTION......................................................................................................................... 1 1.1 Overview of language visualisation ....................................................................................... 2 1.1.1 Multimodal output............................................................................................................ 2 1.1.2 Animation......................................................................................................................... 3 1.1.3 Intelligent ......................................................................................................................... 4 1.2 Problems in language visualisation ........................................................................................ 4 1.3 Objectives of this research...................................................................................................... 5 1.4 Outline of this thesis............................................................................................................... 5 2. APPROACHES TO MULTIMODAL PROCESSING...............................................................8
    [Show full text]