AComputational Grammar for Swedish
                       Noun Phrases in a Natural Language
                              Interface to Databases
                                    Master’s thesis
                                 Department of Linguistics
                                  University of Uppsala
                                      Sweden
                                    Karl Lundstedt
                      Academic supervisor: Anna S˚agvall Hein
                      Industrial supervisor: Gregor Jonsson,
                                   Dialogue Technologies AB
                      Uppsala, 10th December 2002.
          Abstract
          This thesis describes a computational grammar that analyzes Swedish noun phrases, and maps
          them to semantic representations. The noun phrase grammar is a component of the Swedish
          version of the Phoenix natural language interface to databases system which maps natural language
          questions to corresponding SQL queries. The coverage of the grammar is determined by the noun
          phrase part of a grammar specification for Swedish questions. Beside grammatical literature, a
          Wizard of Oz dialogue corpus has been used to investigate the grammatical phenomena supported
          by the grammar specification. An evaluation corpus has been compiled by randomly selecting
          questions from Internet FAQ:s about cellular phones, Internet banking and broadband connection.
          From these questions, noun phrases were extracted and supplied to the grammar. After manually
          examining the resulting semantic representations, it was found that 83% of a total of 254 noun
          phrases were given all the intended semantic representations, and only the intended semantic
          representations.
          Keywords: Natural language interfaces to databases, Computational grammars, Noun phrases,
            Question–Answering systems, Swedish grammar
                               i
                     Acknowledgment
                     I would like to thank Gregor Jonsson, Joakim Ingers, Christopher Dahr´en and Bjorn Als´en at
                                                                                                       ¨
                     Dialogue Technologies for making this thesis possible; professor Anna S˚agvall–Hein for guidance
                     and fruitful discussions; Hubert Lehmann and Thomas Haag at linguatec Entwicklung & Services
                     GmbHfor providing professional computational linguistic knowledge and skills; Peter Cedermark
                     whohasdevelopedtheSwedishnounandadjectivemorphologyusedinthenounphrasegrammar1;
                     Jorg Tiedemann for giving me valuable comments on the evaluation chapter; and Arne Jonsson
                      ¨                                                                                      ¨
                     at NLPLAB, Linkoping University for letting me use their dialogue corpus. I would also like to
                                       ¨
                     thank my wonderful girlfriend Elin for putting up with me in the not always so smooth course of
                     writing.
                       1Cedermark, P. Swedish noun and adjective morphology in a natural language interface to databases. Depart-
                     ment of Linguistics, Uppsala. (in progress)
                                                                   ii
                       Contents
                       1 Introduction                                                                                             1
                       2 Background                                                                                               3
                           2.1   Question Answering systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .         3
                                 2.1.1   Natural Language Interfaces to highly organized data          . . . . . . . . . . . .    3
                                 2.1.2   Q/Asystems finding answers in large collections of free text . . . . . . . . .            5
                                 2.1.3   Advantages and disadvantages with NLIDBs . . . . . . . . . . . . . . . . .               6
                           2.2   The Phoenix NLIDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .          7
                                 2.2.1   System overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        8
                                 2.2.2   The Phoenix parser and the XLDG grammar formalism . . . . . . . . . . .                  8
                       3 Method                                                                                                 10
                           3.1   The Swedish grammar specification . . . . . . . . . . . . . . . . . . . . . . . . . . .          10
                           3.2   Empirical material – Wizard of Oz corpus . . . . . . . . . . . . . . . . . . . . . . .          10
                           3.3   Theoretical material – grammatical literature . . . . . . . . . . . . . . . . . . . . .         11
                                 3.3.1   Traditional Swedish grammars . . . . . . . . . . . . . . . . . . . . . . . . .          11
                                 3.3.2   Swedish computational grammars . . . . . . . . . . . . . . . . . . . . . . . .          11
                       4 The NP grammar                                                                                         12
                           4.1   Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    12
                           4.2   Pre-posed attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      14
                                 4.2.1   Adjectival phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      14
                                 4.2.2   Measure phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       14
                           4.3   Post-posed attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     15
                                 4.3.1   Prepositional phrases     . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   15
                                 4.3.2   Coordination of prepositional phrases       . . . . . . . . . . . . . . . . . . . . .   17
                           4.4   NPlevel rules – common nouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . .          18
                                 4.4.1   Syntactic categories of the NP      . . . . . . . . . . . . . . . . . . . . . . . . .   19
                                 4.4.2   Selecting correct specifier . . . . . . . . . . . . . . . . . . . . . . . . . . . .      19
                                 4.4.3   Prepositional attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      20
                                 4.4.4   Adjectival attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     22
                                 4.4.5   Quantity attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     23
                                 4.4.6   Definite attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      24
                                 4.4.7   Naked nouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       25
                           4.5   Proper nouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      27
                           4.6   Coordination of NP:s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      28
                                 4.6.1   Introduction     . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  28
                                 4.6.2   Restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    31
                                 4.6.3   Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     33
                                                                            iii