A vietnamese text based conversational agent

A Vietnamese Text-based Conversational Agent Nguyen Quoc Dai Faculty of Information Technology University of Engineering and Technology Vietnam National University, Hanoi Supervised by Dr Pham Bao Son A thesis submitted in fulfillment of the requirements for the degree of Master of Science in Computer Science November 2011 ORIGINALITY STATEMENT ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at University of Engineering and Technology (UET/Coltech) or any other educational institution, except where due acknowledgement is made in the thesis Any contribution made to the research by others, with whom I have worked at UET/Coltech or elsewhere, is explicitly acknowledged in the thesis I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.’ rd Hanoi, November 23 , 2011 Signed i ABSTRACT The first step that a question answering system must perform is to transform an input question into an intermediate representation Most published works so far use rule-based approaches to realize this transformation in question answering sys-tems Nevertheless, in existing rule-based approaches, manually creating the rules is error-prone and expensive in time and effort In this thesis, we focus on introduc-ing a rule-based approach that offers an intuitive way to create compact rules for extracting intermediate representation of input questions Experimental results are promising where our system achieves reasonable performance and demonstrate that it is straightforward to adapt to new domains and languages More importantly, this thesis introduces a Vietnamese text-based conversational agent architecture on specific knowledge domain which is integrated in a question answer-ing system When the question answering system fails to provide answers to user input, our conversational agent can step in to interact with users to provide answers to users Experimental results are promising where our Vietnamese text-based con-versational agent achieves positive feedback in a study conducted in the university academic regulation domain Publications: ? Dai Quoc Nguyen, Dat Quoc Nguyen and Son Bao Pham A Vietnamese Text-based Conversational Agent In Proc of The 25th International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems (IEA/AIE 2012), Springer-Verlag LNAI, pp 699-708 ? Dai Quoc Nguyen, Dat Quoc Nguyen and Son Bao Pham A Semantic Approach for Ques-tion Analysis In Proc of The 25th International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems (IEA/AIE 2012), Springer-Verlag LNAI, pp 156-165 ? Dat Quoc Nguyen, Dai Quoc Nguyen and Son Bao Pham Systematic Knowledge Acquisition for Question Analysis In Proc of the 8th International Conference on Recent Advances in Natural Language Processing (RANLP 2011), ACL Anthology, pp 406-412 ii iii ? Dai Quoc Nguyen, Dat Quoc Nguyen, Khoi Trong Ma and Son Bao Pham Automatic Ontology Construction from Vietnamese text In Proceedings of the 7th International Conference on Natural Language Processing and Knowledge Engineering (NLPKE’11), IEEE, pp 485-488 ? Dat Quoc Nguyen, Dai Quoc Nguyen, Son Bao Pham and Dang Duc Pham Ripple Down Rules for Part-Of-Speech Tagging In Proc of 12th International Conference on Intelligent Text Processing and Computational Linguistics (CICLING 2011), Springer-Verlag LNCS, part I, pp 190-201 ? Dai Quoc Nguyen, Dat Quoc Nguyen and Son Bao Pham A Vietnamese question answering system In Proceedings of the 2009 International Conference on Knowledge and Systems Engineer-ing (KSE 2009) , IEEE CS, pp 26 32 ACKNOWLEDGEMENTS First and foremost, I would like to express my deepest gratitude to my supervisor, Dr Pham Bao Son, for his patient guidance and continuous support throughout the years He always appears when I need help, and responds to queries so helpfully and promptly I would like to give my honest appreciation to my younger brother, Nguyen Quoc Dat, for his great support I would like to specially thank Prof Bui The Duy and my colleagues for their help through my time at Human Machine Interaction Laboratory, UET/Coltech I sincerely acknowledge the Vietnam National University, Hanoi, Toshiba Founda-tion Scholarship, and especially Dr Pham Bao Son for supporting finance to my master study Finally, this thesis would not have been possible without the support and love of my mother and my father Thank you! iv To my family ~ v Table of Contents Introduction 1.1 1.2 1.3 A Semantic Approach for Question Analysi A Vietnamese Text-based Conversational A Thesis Organisation Literature review 2.1 Text-based conversational agents 2.1.1 2.1.2 2.2 2.3 FrameScript Scripting Language Question answering systems Our Question Answering System Architecture 3.1 Vietnamese Question Answering System 3.1.1 3.1.2 Using FrameScript for question analysis 3.2.1 3.2.2 3.2.3 3.2 Text-based Conversational Agent for Vietnamese 4.1 4.2 4.3 vi Overview of architecture Determining separate contexts Identifying hierarchical contexts TABLE OF CONTENTS Evaluation and Discussion 5.1 5.2 5.3 Conclusion A Scripting patterns for English question analysis B Definitions of question-class types C Definitions of question-structures Experimental results for Vietnamese text-based conversa Question Analysis for English Discussion List of Figures 2.1 O’Shea et al.’s conversational agent framework 2.2 Aqualog’s architecture 3.1 Architecture of our question answering system 3.2 Architecture of the natural language question analysis componen using FrameScript 4.1 Architecture of our Vietnamese text-based conversational agent viii 32 Table 5.5: Number of rules with conditional re Question-structure type UnknTerm ThreeTerm ThreeTerm Combine As the intermediate representation of our system is different to AquaLog and there is no common test set available, it is difficult to directly compare our approach with Aqualog on the English domain To demonstrate that our approach could be applied to a new open domain, we use the above 52 rules to test the data of 500 questions from TREC 10 Table 5.6 shows the number of correctly analyzed questions corresponding with each question-structure type Table 5.6: Number of questions corresponding with each question-structure type Question-structure type Definition UnknTerm UnknRel Normal ThreeTerm And Table 5.7 gives the sources of errors for 259 incorrect cases It clearly shows that most errors come from unexpected structures This could be rectified by adding more rules, especially when we construct a larger variety of question structure types from a bigger training data such as 5500 questions (Li and Roth, 2002a) Table 5.7: Error results Reason Have special characters (such as /’s) and abbreviations Not have compatible patterns Semantic error in elements of the intermediate representation This experiment is indicative of the ability in using our system to quickly build rules for a new domain We believe that our approach could be applied to a new http://cogcomp.cs.illinois.edu/Data/QA/QC/TREC_10.label 5.3 Discussion language because creating the rules manually for question analysis is a language independent process 5.3 Discussion Because constructing rules depends on the identification of super-contexts and their sub-contexts in Vietnamese, so it causes difficulties in designing the hierarchy of contexts Consequently, we want to simplify this designing phase according to the process of semantic knowledge acquisition We built additional scripts by using same process shown in section 3.2 to detect noun phrases, question phrases and relation phrases or semantic constraints between them for Vietnamese Using these scripts, we constructed pattern expressions and got the suitable phrases from response ex-pressions These phrases actually are keywords which may be used as patterns of rules in the hierarchical contexts In addition, our Vietnamese text-based conversational agent is integrated with our ontology-based Vietnamese question answering system to form a general system Our goal is to retrieve the necessary information from user’s utterance to support our Vietnamese question answering system in providing answers to users We con-sider the process that the Answer retrieval component similarly measures between elements of the intermediate representation of user’s question and the ontology’s elements In case of ambiguity for the similarity among ontology’s elements is still present, the system will interacts with the users by presenting different options to get the correct ontology’s elements In this process, we can construct the supplemental scripts to solve ambiguities from ontology knowledge base Using these scripts, we can retrieve the suitable elements from ontology through the conversational contexts structured based on the given ontology Chapter Conclusion In this thesis, we firstly presented, in chapter 3, our first Ontology-based Vietnamese question answering system with two components The first component takes the user’s question and transforms it into an intermediate representation in a compact form The intermediate representation used in our system comprises of a question-structure and one or more query-tuples in the format of (questionstructure, question-class, Term1, Relation, Term2, Term3), in which Term1 represents a concept (object class), Term 2, and Term3, if exist, represent entities (objects), Relation (property) is a semantic constraint between terms in the question The second component maps terms and relations in the query-tuple to concepts, instances and relations in the ontology to generate a semantic answer Obviously, we spent a large amount of time for writing grammar rules to analyze input questions and did realize difficulties in controlling interactions between these rules Consequently, also in chapter 3, we introduced a rule-based approach for con-verting a natural language question into an intermediate representation in a question answering system Our system utilizes FrameScript to help users to create intuitive and compact rules for extracting elements of the intermediate representation We constructed rules including patterns and associated responses, in which pattern is used to match user’ questions and its corresponding response as output is sent to return the intermediate representation of question Experimental results of our system on a wide range of questions are promising with reasonable performance (in chapter 5) We believe our approach can be applied to question answering in open domain against text corpora that requires an analysis to turn an input question to an explicit representation of some sort Our 34 35 method could be combined nicely with the processing of annotating corpus and it is straightforward to apply for a new domain and a new language We will extend our system in the future to assist the rule creation process on a wide range of questions in open domain and to improve the generalization capability of existing rules Importantly, in this thesis, we proposed a Vietnamese text-based conversational agent architecture as backup component integrated with our Vietnamese question answering system to form a general system (in chapter 4) We focused on presenting an approach to construct the hierarchical contexts consisting of organized rules over a specific knowledge domain There are two steps to construct the conversational contexts: the first step to identify the transformations from a context to other con-texts, and the second step to organize these contexts into a hierarchy to handle unexpected inputs Our contribution is to provide the suitable information related to users’ statements and to retrieve the necessary knowledge to support our question answering system in providing answers The experimental results (as described in chapter 5) are promising, with positive evaluation from users for our Vietnamese text-based conversational agent To the best of our knowledge, this is the first Vietnamese text-based conversational agent to enable users to interact with the system via a natural language interface In the future, our text-based conversational agent will be extended not only to communicate with users but also to get the necessary information related to ontology knowledge base from input utterances We will build scripts to resolve the ambiguity between elements of ontology such as the similarity of string names among classes or instances in the ontology The constructed scripts would be utilized to generate options in order to obtain terms from conversational interactions with users Appendix A Scripting patterns for English question analysis ==> ^ ^ ^ ^ [ ( == is or == are) > Definition, (Definition, What, ?, ?, 3, ?) | ^ ^ UnknTerm, (UnknTerm, What, ?, 2, 3, ?) ] ==> ^ ^ ^ ^ [ ( == is or == are) > Definition, (Definition, Who, ?, ?, 3, ?) | ^ ^ UnknTerm, (UnknTerm, Who, ?, 2, 3, ?) ] ==> ^ ^ [ UnknTerm, (UnknTerm, Which, ?, 2, 3, ?) ] ==> ^ ^ [ UnknRel, (UnknRel, Entity, 1, ?, 3, ?) ] ==> ^ ^ [ UnknRel, (UnknRel, List, 1, ?, 3, ?) ] ==> ^ ^ ^ [ Normal, (Normal, Entity, 1, 2, 3, ?) ] ==> ^ ^ ^ [ Normal, (Normal, List, 1, 2, 3, ?) ] ==> 36 Appendix A ^ ^ ^ [ Affirm, (Affirm, YesNo, 2, 3, 4, ?) ] ==> ^ ^ ^ [ Affirm, (Affirm, YesNo, 2, 1, 3, ?) ] { | } ==> ^ ^ ^ ^ [ ThreeTerm, (ThreeTerm, List, 1, 2, 3, 6) ] ==> [ ^ ^ ^ ^ ThreeTerm, (ThreeTerm, Entity, 1, 4, 3, 5) ] ==> [ ^ ^ ^ ^ And, (UnknTerm, Who, ?, 2, 3, ?), (UnknTerm, Who, ?, 5, 6, ?) ] ==> ^ ^ ^ ^ ^ ^ [ And, (Normal, Entity, 1, 2, 3, ?), (Normal, Entity, 1, 5, 6, ?) ] ==> ^ ^ ^ ^ [ Or, (UnknTerm, Who, ?, 2, 3, ?), (UnknTerm, Who, ?, 2, 6, ?) ] ==> ^ ^ ^ ^ [Clause, (UnknTerm, What, ?, 2, ?, ?), (Normal, Which, 3, 5, 6, ?) ] Appendix B Definitions of question-class types In our approaches, question is classified into one of the following classes of HowWhy, YesNo, What, When, Where, Who, Many, ManyClass, List, and Entity To iden-tify question categories, we specify a number of JAPE grammars using NounPhrase annotations and the question-word information identified by the preprocessing mod-ule Obviously using this method in question-phrases detection phase will result in ambiguity when a question belongs to multiple categories We allow for this and resolve the ambiguity in the semantic analysis module A question referring causes or methods by containing the word re-annotated by single TokenVn annotation such as t⁄i saowhy , or v… saowhy , or th‚ n ohow , or l nh÷ th‚ n ohow , , will be classified to HowWhy (in English, it appertains types of Why-questions and How is/are questions) A question requiring true or false answer by holding the word re-covered by single TokenVn annotations such as câ óng l is that , or câ ph£i l is this , or ph£i khỉngare there , or óng khỉngare those , , will be categorized to YesNo class A question which refers to something in consisting of the word c¡i g… what , or l g…what , or l nhœng c¡i g…what , , are classified to What class In English, this question type is What is/are-question-like A question containing the word relabelled by single TokenVn annotation such as n owhen , or v o thíi gian n owhich time , or lóc n owhen , or ng y n owhich date , , will belong to When class 38 Appendix B A question which requires answers about location in consisting of word ð nìi n owhere , or l ð nìi ¥uwhere , or ð chØ n owhere , , appertains to Where category (in English, this question type is kind of Where-questions) A question will be categorized to Who class, if it contains the word indicating answer referring to a person such as l nhœng who , or l ng÷íi n owho , or nhœng aiwho , A question expecting the answer about number will belong to Many class (in English, these questions are How much/many is/are-questions) This question type holds the word like bao nhi¶uhow muchjmany , or l bao nhi¶uhow muchjmany , or sŁ l÷ỉnghow many , A question will appear in ManyClass class, if it contains the word like bao nhiảuHowmany , or s lữổngNumberof followed by a noun phrase (in English, this type is the same kind of How many NounPhrase-question) A question will appertain in Entity category if it holds a noun phrase followed by the word n owhich or g…what (in English, this kind of question belongs to set ofwhich/what Noun Phrase questions such as: which students, what class, ) A question will categorized to List class if it contains the word referring commands such as: cho bi‚tgive , ch¿ rashow , k” ratell , t…mfind , li»t k¶list followed by a noun phrase Appendix C Definitions of question-structures We define question structures: Normal, UnknTerm, UnknRel, Definition, Compare, ThreeTerm, Clause, Combine, And, Or, Affirm, Affirm_3Term, Affirm_MoreTuples as following: A question will have question-structure of Normal if it has only one querytuple, and the query-tuple’s Term3 is missing A question will have question-structure of UnknTerm if it has only one query-tuple, and the query-tuple lacks Term1 and Term3 A question will have UnknRel question-structure if it has only one querytuple in the lack of Relation and Term3 A question will have Definition question-structure if it has only one querytuple lacking all of Term1, Relation and Term3 For example, the question Nguy„n QuŁc ⁄i l ai? ( who is Nguyen Quoc Dai? ) has query-tuple ( Definition , Who , ? , ? , Nguy„n QuŁc ⁄i NguyenQuocDai , ? ) If a question belongs to one of three question-structure types Normal, Unkn-Rel and UnknTerm and appears in YesNo category, the question will have question-structure of Affirm For example, the question Tỗn t⁄i sinh vi¶n l Trƒn B…nh Giang ph£i khỉng? ( Is there a student who Tran Binh Giang is? ) has query-tuple ( Affirm , YesNo , sinh vi¶nstudent , ? , Trƒn B…nh GiangT ran Binh Giang , ? ) 40 Appendix C A question will have ThreeTerm question-structure if it has only one querytuple, and one or more of the query-tuple’s Term and Relation is missing For instance, the question nhœng l gi£ng vi¶n cıa khoa cỉng ngh» thỉng tin cıa tr÷íng ⁄i håc Cỉng Ngh»? ( which lecturers are from faculty of Informa-tion Technology of College of Technology? ) has query-tuple ( ThreeTerm , Who gi£ng vi¶n is lecturer , khoa cỉng ngh» thỉng tin ,?,l faculty ofInformation T echnology , tr÷íng ⁄i håc Cỉng Ngh»College of T echnology ) If a question has question-structure of ThreeTerm and appears in YesNo cat-egory, it will have Affirm_3Term question-structure Given the question s lữổng sinh viản hồc lỵp khoa håc m¡y t‰nh l 45 ph£i khỉng? ( is that the number of students studying in computer science class is 45? ) has query-tuple ( Affirm_3Term, ManyClass, sinh viảnstudent, hồcstudy, lợp khoa hồc mĂy tnh computerscienceclass , 45 ) A question will have question-structure of Compare if it belongs to one of three question-structure types Normal, UnknRel and UnknTerm, and contains a comparing-phrase which is detected by preprocessing module; the query-tuple’s Term3 is used to hold this comparison information For example, the question sinh vi¶n n o câ i”m trung b…nh cao nh§t khoa cỉng ngh» thỉng tin? ( which student has highest grade point average in faculty of Information Technology? ) has query-structure of Compare and query-tuple (UnknTerm , Who , ? , i”m trung b…nhgrade point average , khoa cæng ngh» thæng tin faculty of Information T echnology , cao nh§t highest ) If a question contains either token v and or m and / ho°cor , it will have two or more query-tuples corresponding with And/Or question-structure However, with some question like Ph⁄m øc «ng hồc trữớng n o v ữổc hữợng dÔn bi ai? ( Which university does Pham Duc Dang study and who advises him? ), it contains v and , but it has question-structure of Or and two following query-tuples: ( Normal , Entity , trữớngschool , hồcstudy , Phm ức ôngP ham Duc Dang , ? ) and ( UnknTerm , Who , ? , hữợng dÔnguide , Phm ức ôngP ham Duc Dang , ? ) 42 Appendix C If a question appearing in YesNo class belongs to And or Or questionstructure types, it will have Affirm_MoreTuples question-structure If a question corresponds two query-tuples and returned results of this querytuple is considered as miss-element in remaining query-tuple, it will have question-structure of Clause For example, the question nhœng sinh viản cõ H Ni th hồc lợp n o? ( which classes students coming Hanoi study? ) has question-structure of type Clause and two query-tuples: ( Normal , Entity , lỵpclass , håcstudy , ? , ? ) and ( Normal , ? , sinh vi¶nstudent , câ qu¶has hometown , H NºiHanoi , ? ), without including question-class, We can see that returned results of sub-question represented by second query-tuple indicate missing element of Term2 in the first query-tuple If a composite question is constructed from two or more independent clauses, it will have question-structure of Combine For example, the question câ qu¶ ð H TƠy v sinh viản n o hồc khoa cổng ngh» thæng tin? ( who from Hatay and which student study in faculty of Information Technology? ) has two query-tuples: ( UnknTerm , Who ,? , câ qu¶has hometown , H T¥yHatay , ? ) and ( Normal , Entity , sinh vi¶nstudent , håcstudy , khoa cỉng ngh» thỉng tin faculty ofInformation T echnology , ? ) In each query-tuple, in general, Term1 represents a concept, without cases of Affirm, Affirm_3Term and Affirm_MoreTuples question-structures; Term and Term3, if exist, represent entities (maybe representing concepts in case of Definition question-structure) Bibliography Ion Androutsopoulos, Graeme Ritchie, and Peter Thanisch Natural language interfaces to databases an introduction Natural Language Engineering, 1:29 81, 1995 William W Cohen, Pradeep Ravikumar, and Stephen E Fienberg A comparison of string distance metrics for name-matching tasks In Proceedings of IJCAI-03 Workshop on Information Integration, pages 73 78, 2003 Hammish Cunningham, Diana Maynard, Kalina Bontcheva, and Valentin Tablan GATE: A Frame-work and Graphical Development Environment for Robust NLP Tools and Applications In Proceedings of the 40th Anniversary Meeting of the Association for Computational Linguistics, pages 168 175, 2002 Danica Damljanovic, Valentin Tablan, and Kalina Bontcheva A text-based query interface to owl ontologies In Proceedings of 6th Language Resources and Evaluation Conference, 2008 Christiane D Fellbaum WordNet: An Electronic Lexical Database MIT Press, 1998 A Galea Open-domain surface-based question answering system In Proceedings of the Computer Science Annual Workshop (CSAW), 2003 Arthur C Graesser, Shulan Lu, George Tanner Jackson, Heather Hite Mitchell, Mathew Ventura, Andrew Olney, and Max M Louwerse AutoTutor: A Tutor with Dialogue in Natural Language Behavioral Research Methods, Instruments, and Computers, Vol 36:180 192, 2004 L Hirschman and R Gaizauskas Natural Language Question Answering: The View from here Natural Language Engineering, Vol 7:275 300, 2001 Boris Katz Annotating the world wide web using natural language In Proceedings of the 5th RIAO Conference on Computer Assisted Information Searching on the Internet - RIAO 1997, pages 136 159, 1997 Boris Katz, Gary C Borchardt, and Sue Felshin Natural language annotations for question answering In Proceedings of the 19th International Florida Artificial Intelligence Research Society Conference, pages 303 306, 2006 43 44 Bibliography Anton Leuski, Ronakkumar Patel, David Traum, and Brandon Kennedy Building Effective Question Answering Characters In Proc of the 7th SIGdial Workshop on Discourse and Dialogue, pages 18 27, 2006 Xin Li and Dan Roth Learning Question Classifiers In Proceedings of the 19th International Conference on Computational Linguistics, volume of COLING ’02, pages 7, 2002a Xin Li and Dan Roth Learning question classifiers In Proceedings of the 19th international conference on Computational linguistics - Volume 1, COLING ’02, pages Association for Computational Linguistics, 2002b Yuhua Li, David McLean, Zuhair A Bandar, James D O’Shea, and Keeley Crockett Sentence similarity based on semantic nets and corpus statistics IEEE Trans on Knowl and Data Eng., 18:1138 1150, 2006 Vanessa Lopez, Victoria Uren, Enrico Motta, and Michele Pasin Aqualog: An ontology-driven question answering system for organizational semantic intranets Web Semantics: Science, Ser-vices and Agents on the World Wide Web, 5(2):72 105, 2007 Matthew McGill, Claude Sammut, J Westendorp, and W Kadous Framescript: A multi-modal scripting language In Copyright c 2003-2008 The School of Computer Science and Engineer-ing, UNSW Matthew McGill, Claude Sammut., 2003 Anh Kim Nguyen and Huong Thanh Le Natural language interface construction using seman-tic grammars In Proceedings of the 10th Pacific Rim International Conference on Artificial Intelligence, pages 728 739, 2008 Dai Quoc Nguyen, Dat Quoc Nguyen, and Son Bao Pham A vietnamese question answering system In Proceedings of the 2009 International Conference on Knowledge and Systems Engineering, pages 26 32, 2009 Dat Quoc Nguyen, Dai Quoc Nguyen, and Son Bao Pham Systematic knowledge acquisition for question analysis In Proceedings of 8th International Conference on Recent Advances in Natural Language Processing, (In press), September, 2011 K O’Shea, Z Bandar, and K Crockett A novel approach for constructing conversational agents using sentence similarity measures In Proc of WCE, volume 1, 2008 K O’Shea, Z Bandar, and K Crockett A conversational agent framework using semantic analysis International Journal of Intelligent Computing Research (IJICR), 1(1):23 32, 2010 Dang Duc Pham, Giang Binh Tran, and Son Bao Pham A hybrid approach to vietnamese word segmentation using part of speech tags In Proceedings of the 2009 International Conference on Knowledge and Systems Engineering, pages 154 161, 2009 Bibliography T.T Phan and T.C Nguyen Question semantic analysis in vietnamese qa system In Edited book "Advances in Intelligent Information and Database Systems" of The 2nd Asian Conference on Intelligent Information and Database Systems (CIIDS2010), pages 29 40, 2010 Ana-Maria Popescu, Oren Etzioni, and Henry Kautz Towards a theory of natural language interfaces to databases In Proceedings of the 8th international conference on Intelligent user interfaces, IUI ’03, pages 149 157, 2003 Claude Sammut Claude sammut: Managing context in a conversational agent In Electronic Transactions on Artificial Intelligence, volume 5, pages 189 202, 2001 Ashish Kumar Saxena, Ganesh Viswanath Sambhu, Saroj Kaushik, and L Venkata Subramaniam Iitd-ibmirl system for question answering using pattern matching, semantic type and semantic category recognition In Proceedings of The Sixteenth Text REtrieval Conference, 2007 Eriks Sneiders Automated question answering using question templates that cover the conceptual model of the database In Proceedings of the 6th International Conference on Applications of Natural Language to Information Systems-Revised Papers, NLDB ’02, pages 235 239, 2002 Niculae Stratica, Leila Kosseim, and Bipin C Desai Nlidb templates for semantic parsing In Pro-ceedings of the 8th International Conference on Applications of Natural Language to Information Systems, pages 235 241, 2003 Marjorie Templeton and John Burger Problems in natural-language interface to dbms with exam-ples from eufid In Proceedings of the first conference on Applied natural language processing, pages 16, 1983 David R Traum Talking to Virtual Humans: Dialogue Models and Methodologies for Embodied Conversational Agents In ZiF Workshop, pages 296 309, 2006 M Vargas-Vera and E Motta An ontology-driven similarity algorithm Technical report, Knowledge Media Institute, The Open University, 2004 Richard S Wallace A.l.i.c.e artificial http://www.alicebot.org/about.html, 2001 intelligence foundation, inc Available: David L Waltz An english language question answering system for a large relational database Commun ACM, 21:526 539, July 1978 Joseph Weizenbaum Eliza - a computer program for the study of natural language communication between man and machine Commun ACM, pages 23 28, 1983 Min Wu, Xiaoyu Zheng, Michelle Duan, Ting Liu, and Tomek Strzalkowski Question answering by pattern matching, web-proofing, semantic form proofing In Proceedings of the Twelfth Text REtrieval Conference (TREC 2003), pages 578 585, 2003 46 Bibliography Copyright c 2011 by Dai Quoc Nguyen Printed and bound by Dai Quoc Nguyen ... Chapter Text- based Conversational Agent for Vietnamese In this chapter, we will describe a Vietnamese text- based conversational agent (CA) architecture and focus on creating rules manually in... interfaces to databases A natural language interface to a database (NLIDB) is a system that allows the users to access information stored in a database by typing questions using natural language... 2.1 Text- based conversational agents 2.1.1 Using keywords for pattern matching ELIZA (Weizenbaum, 1983) was one of the earliest text- based conversational agents based on a simple pattern matching