PhD thesis - School of Informatics - University of Edinburgh

More documents

Recommendations

Info

Chapter 5. Parsing English Inclusions 131 Group Mean StDev Inclusion set 0.008007 0.002140 Random set 0.008683 0.001781 Table 5.4: Means and standard deviations of frequency profiles for the inclusion and the stratified random set. 5.3.5 Error Analysis The aim of this error analysis is to examine the exact reasons for why the inclusion entity model yields an improved performance over the baseline model when parsing sentences containing English inclusions. The error analysis is limited to 100 sentences selected from the inclusion set parsed with both the baseline and the inclusion entity model. This sample contains 109 English inclusions, five of which are false positives, i.e., the output of the English inclusion classifier is incorrect. The precision of the classifier in recognising multi-word English inclusions is therefore 95.4% for this TIGER sample. Before analysing the errors in the parsing output of the inclusion set in detail, it is worthwhile examining typical gold standard phrase categories of the multi-word English inclusions. Phrase category Frequency Example PN 91 The Independent CH 10 Made in Germany NP 4 Peace Enforcement CNP 2 Botts and Company – 2 Chief Executives Table 5.5: Gold standard phrasal categories of English inclusions.
Chapter 5. Parsing English Inclusions 132 5.3.5.1 Gold Standard Phrase Categories Table 5.5 lists the different types of phrase categories surrounding the 109 multi-word English inclusions in the error analysis sample and their frequency. 6 The last column lists a typical example for each category. The figures illustrate that the majority of multi-word English inclusions are contained in a proper noun (PN) phrase, including names of companies, political parties, organisations, films, books, newspapers, etc. The components of PN phrases tend to be marked with the grammatical function PNC, proper noun component (Brants et al., 2002). A less frequent phrasal category of En- glish inclusions is chunk (CH) which tends to be used for slogans, quotes or expressions like Made in Germany. The components of CH phrases are annotated with a grammatical function of type UC (unit component). Even in this small sample, phrase category annotations of English inclusions as either PN or CH, and not the other, can be mislead- ing. For example, the organisation Friends of the Earth is annotated as PN, whereas another organisation International Union for the Conservation of Nature is marked as CH in the gold standard. The latter is believed to be an inconsistency in the annotation and should have been marked as PN as well. The phrase category of an English inclusion with the syntactic function of a noun phrase which is neither a PN nor a CH is annotated as NP (noun phrase). One example is Peace Enforcement which is not translated into German and used rather like a buzzword in a sentence on UN missions. In this case, the POS tag of its individual tokens is NN (noun). The fact that this expression is not German is therefore lost in the gold standard annotation. Another example of an English inclusion NP in the gold standard is Framingham Heart Study which could arguably be of phrase category PN. Furthermore, the sample contains an example of phrase category CH, Shopping Mall, an English noun phrase. The least frequent type of phrase category used for English inclusions is CNP. In this sample, this category marks a company names made up of a conjunction, for example Botts and Company. The POS tags of the coordinated sisters are NE (named entity) and the English coordinated conjunction and is tagged as KON. Finally, there are also two cases, where the English inclusion itself is not contained in a phrase category. One of them is Chief Executives which is clearly an NP. These are believed to be annotation errors. 6 All phrase category (node) labels and grammatical function (edge) labels occurring in the TIGER treebank annotation are listed and defined in Appendix C.
Page 1 and 2:
Automatic Detection of English Incl
Page 3 and 4:
these parsers with the annotation-f
Page 5 and 6:
Declaration I declare that this the
Page 7 and 8:
3.3.5 Post-processing Module . . .
Page 9 and 10:
A.2.2 Kappa Coefficient . . . . . .
Page 11 and 12:
5.6 Average relative token frequenc
Page 13 and 14:
3.16 Most frequent English inclusio
Page 15 and 16:
Chapter 1. Introduction 2 siderable
Page 17 and 18:
Chapter 1. Introduction 4 Chapter 3
Page 19 and 20:
Chapter 1. Introduction 6 1.1 Relat
Page 21 and 22:
Chapter 2. Background and Theory 8
Page 23 and 24:
Page 25 and 26:
Page 27 and 28:
Page 29 and 30:
Page 31 and 32:
Page 33 and 34:
Page 35 and 36:
Page 37 and 38:
Page 39 and 40:
Page 41 and 42:
Page 43 and 44:
Page 45 and 46:
Page 47 and 48:
Page 49 and 50:
Page 51 and 52:
Page 53 and 54:
Page 55 and 56:
Page 57 and 58:
Page 59 and 60:
Chapter 3 Tracking English Inclusio
Page 61 and 62:
Chapter 3. Tracking English Inclusi
Page 63 and 64:
Page 65 and 66:
Page 67 and 68:
Page 69 and 70:
Page 71 and 72:
Page 73 and 74:
Page 75 and 76:
Page 77 and 78:
Page 79 and 80:
Page 81 and 82:
Page 83 and 84:
Page 85 and 86:
Page 87 and 88:
Page 89 and 90:
Page 91 and 92:
Page 93 and 94: Chapter 3. Tracking English Inclusi
Page 113 and 114: Chapter 4 System Extension to a New
Page 115 and 116: Chapter 4. System Extension to a Ne
Page 129 and 130: Chapter 5 Parsing English Inclusion
Page 131 and 132: Chapter 5. Parsing English Inclusio
Page 143: Chapter 5. Parsing English Inclusio
Page 159 and 160: Chapter 6 Other Potential Applicati
Page 161 and 162: Chapter 6. Other Potential Applicat
Page 187 and 188: Chapter 7 Conclusions and Future Wo
Page 189 and 190: Chapter 7. Conclusions and Future W
Page 191 and 192: Appendix A. Evaluation Metrics and
Page 193 and 194: Appendix A. Evaluation Metrics and
Page 195 and 196:
Appendix A. Evaluation Metrics and
Page 197 and 198:
Appendix A. Evaluation Metrics and
Page 199 and 200:
Appendix B. Guidelines for Annotati
Page 201 and 202:
Page 203 and 204:
Page 205 and 206:
Appendix C TIGER Tags and Labels C.
Page 207 and 208:
Appendix C. TIGER Tags and Labels 1
Page 209 and 210:
Appendix C. TIGER Tags and Labels 1
Page 211 and 212:
Bibliography 198 Andersen, G. (2005
Page 213 and 214:
Bibliography 200 Bresnan, J. (2001)
Page 215 and 216:
Bibliography 202 Damashek, M. (1995
Page 217 and 218:
Bibliography 204 Finkel, J., Dingar
Page 219 and 220:
Bibliography 206 Hachey, B., Alex,
Page 221 and 222:
Bibliography 208 Kirkness, A. (1984
Page 223 and 224:
Bibliography 210 and Technology (In
Page 225 and 226:
Bibliography 212 Poplack, S. (1988)
Page 227 and 228:
Bibliography 214 Sokol, D. K. (2000
Page 229:
Bibliography 216 Yang, W. (1990). A
show all

PhD thesis - School of Informatics - University of Edinburgh

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?