Developing Enterprise Chatbots Learning Linguistic Structures
A chatbot is expected to be capable of supporting a cohesive and coherent conversation and be knowledgeable, which makes it one of the most complex intelligent systems being designed nowadays. Designers have to learn to combine intuitive, explainable lang
- PDF / 19,719,455 Bytes
- 566 Pages / 439.43 x 683.15 pts Page_size
- 1 Downloads / 248 Views
Developing Enterprise Chatbots Learning Linguistic Structures
Developing Enterprise Chatbots
Boris Galitsky
Developing Enterprise Chatbots Learning Linguistic Structures
Boris Galitsky Oracle (United States) San Jose, CA, USA
Supplementary material and code is available at https://github.com/bgalitsky/relevancebased-on-parse-trees ISBN 978-3-030-04298-1 ISBN 978-3-030-04299-8 https://doi.org/10.1007/978-3-030-04299-8
(eBook)
Library of Congress Control Number: 2019932803 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword
We launched Ask Jeeves in April 1997, with Yahoo!, Alta Vista, and Excite all already in the market. We wanted to set ourselves apart from conventional search engines with a special interface. Our team was building a library of “knowledge capsules,” snapshots of answers to the most popular questions. For a user question, we were trying to find the most similar one indexed in our system and return it along with answers. If a question was not included in our index, then we would fall back to a more general search. Ask Jeeves was more relevant for users overwhelmed with pages of results stemming from a simple search. Most queries were consumer-oriented – asking for the best restaurants, sports scores, or pictures of Pamela Anderson – while others required the kind of information associated with urgency and specialized knowledge, such as “How to get rid of flu?”. This approach proved quite effective. People enjoyed the direct, personalized navigation, and saw themselves as Ask Jeeves loyalists. As Google came to dominate the search engine market, search results became less direct and personalized, and more optimized for advertise
Data Loading...