Academia.eduAcademia.edu

Outline

Title
Abstract
Figures
Introduction
System Design
Feature Analysis
Experimental Results
Sentiment Analysis
Conclusions
References
All Topics
Art
Digital Humanities

A Systemic Functional Approach to Automated Authorship Analysis

Profile image of Moshe KoppelMoshe Koppel

2013

https://doi.org/10.1002/ASI.20553
visibility

…

description

21 pages

link

1 file

Abstract

Most text analysis and retrieval work to date has focused on the topic of a text; that is, what it is about. However, a text also contains much useful information in its style, or how it is written. This includes information about its author, its purpose, feelings it is meant to evoke, and more. This article develops a new type of lexical feature for use in stylistic text classification, based on taxonomies of various semantic functions of certain choice words or phrases. We demonstrate the usefulness of such features for the stylistic text classification tasks of determining author identity and nationality, the gender of literary characters, a text's sentiment (positive/ negative evaluation), and the rhetorical character of scientific journal articles. We further show how the use of functional features aids in gaining insight about stylistic differences among different kinds of texts.

Figures (18)
TABLE 1. The authorship attribution corpus, comprising the chapters in a set of 20 19-century novels. The number of chapters in each book and the average number of words per chapter in each book are as shown.
TABLE 1. The authorship attribution corpus, comprising the chapters in a set of 20 19-century novels. The number of chapters in each book and the average number of words per chapter in each book are as shown.
FIG. 2. Ten-fold cross-validation accuracy for book attribution in 19th-century literature. Baseline (majority class) classification would give 9.6% accuracy. FIG. 1. Ten-fold cross-validation accuracy for authorship attribution in 19th-century literature. Baseline (majority class) classification would give 24% accuracy.
FIG. 2. Ten-fold cross-validation accuracy for book attribution in 19th-century literature. Baseline (majority class) classification would give 9.6% accuracy. FIG. 1. Ten-fold cross-validation accuracy for authorship attribution in 19th-century literature. Baseline (majority class) classification would give 24% accuracy.
We may tentatively examine differences between the tasks based on which functional features seemed to help the most. Examination of the features shows some clear differences among the two tasks, in terms of which sorts of features were most significant for classification, from which we can draw some tentative conclusions. Book discrimination involves a
We may tentatively examine differences between the tasks based on which functional features seemed to help the most. Examination of the features shows some clear differences among the two tasks, in terms of which sorts of features were most significant for classification, from which we can draw some tentative conclusions. Book discrimination involves a
TABLE 2. The top 20 features (by rank sum) for each of book, authorship, and nationality attribution.
TABLE 2. The top 20 features (by rank sum) for each of book, authorship, and nationality attribution.
FIG. 3. Ten-fold cross-validation accuracy for nationality attribution in 19th-century literature. Baseline (majority class) classification would give 54.0% accuracy.
FIG. 3. Ten-fold cross-validation accuracy for nationality attribution in 19th-century literature. Baseline (majority class) classification would give 54.0% accuracy.
TABLE 3. Oppositions from the 15 highest ranked systemic features for each class, with weights taken from the model learned using FW + Com + Mod for Nationality attribution.
TABLE 3. Oppositions from the 15 highest ranked systemic features for each class, with weights taken from the model learned using FW + Com + Mod for Nationality attribution.
TABLE 4. Composition of the corpus of characters’ speeches from Shakespeare. The table shows the number and average total speech length of selected characters of each gender from each play. The text has more detail on the construction of the corpus.
TABLE 4. Composition of the corpus of characters’ speeches from Shakespeare. The table shows the number and average total speech length of selected characters of each gender from each play. The text has more detail on the construction of the corpus.
FIG. 4. Accuracies for various feature sets over the corpus.
FIG. 4. Accuracies for various feature sets over the corpus.
TABLE 5. Oppositions found in the top 20 features from both genders of Shakespearean characters. Note that they are organized categorically, not in rank order.
TABLE 5. Oppositions found in the top 20 features from both genders of Shakespearean characters. Note that they are organized categorically, not in rank order.
FIG. 5. Movie review classification results for using SMO with default parameters and a linear kernel with various feature sets; see text for further details.
FIG. 5. Movie review classification results for using SMO with default parameters and a linear kernel with various feature sets; see text for further details.
TABLE 6. Top 15 features for positive and negative reviews from BoW + App, showing wo (multiplied b’ 10,000 for scaling).
TABLE 6. Top 15 features for positive and negative reviews from BoW + App, showing wo (multiplied b’ 10,000 for scaling).
FIG. 6. Ten-fold cross-validation accuracies for SMO with various feature sets, on the corpus of science articles.
FIG. 6. Ten-fold cross-validation accuracies for SMO with various feature sets, on the corpus of science articles.
TABLE7. Summary of the geology and paleontology journals used in the scientific literature corpus study, giving the number of articles from each journal in the corpus, and the average number of words per article.
TABLE7. Summary of the geology and paleontology journals used in the scientific literature corpus study, giving the number of articles from each journal in the corpus, and the average number of words per article.
TABLE 8. Oppositions from the 20 highest ranked systemic features in geology and paleontology articles, from the model learned using FW + Con + Com + Mod + App.
TABLE 8. Oppositions from the 20 highest ranked systemic features in geology and paleontology articles, from the model learned using FW + Con + Com + Mod + App.
FIG. Al. The CONJUNCTION system (Matthiessen, 1995). Options here are disjunctive; examples of lexical realizations for the leaves are given in italics. Different patterns of CONJUNCTION usage lead to markedly different textual styles. Frequent use of Extension can give a text with high information density which can give a “panoramic” effect of touring through a conceptual landscape, but if done poorly may overwhelm and lose a reader in too many facts. On the other hand, Elaboration can be used to good effect to create textual coherence around a single focused storyline. We note, too, that many of the standard function words traditionally used in computational stylistic studies are types of CONJUNCTION, which fur- ther argues for this system’s importance for stylistic text analysis.
FIG. Al. The CONJUNCTION system (Matthiessen, 1995). Options here are disjunctive; examples of lexical realizations for the leaves are given in italics. Different patterns of CONJUNCTION usage lead to markedly different textual styles. Frequent use of Extension can give a text with high information density which can give a “panoramic” effect of touring through a conceptual landscape, but if done poorly may overwhelm and lose a reader in too many facts. On the other hand, Elaboration can be used to good effect to create textual coherence around a single focused storyline. We note, too, that many of the standard function words traditionally used in computational stylistic studies are types of CONJUNCTION, which fur- ther argues for this system’s importance for stylistic text analysis.
FIG. A2. The MODALITY system networks (Matthiessen, 1995).
FIG. A2. The MODALITY system networks (Matthiessen, 1995).
FIG. A3. Examples of indicator features for various combinations of MODALITY options. Note that not all combinations are realized in the language; note also the ambiguity of some of the indicators.
FIG. A3. Examples of indicator features for various combinations of MODALITY options. Note that not all combinations are realized in the language; note also the ambiguity of some of the indicators.
FIG. A4. Options in the Attitude network, with examples of appraisal adjectives from our lexicon
FIG. A4. Options in the Attitude network, with examples of appraisal adjectives from our lexicon

Related papers

An Empirical Method Exploring a Large Set of Features for Authorship Identification
rim faiz

2016

In this paper, we deal with the author identification issues of the document whose origin is unknown. To overcome these problems, we propose a new hybrid approach combining the statistical and stylistic analysis. Our introduced method is based on determining the lexical and syntactic features of the written text in order to identify the author of the document. These features are explored to build a machine learning process. We obtained promising results by relying on PAN@CLEF2014 English literature corpus. The experimental results are comparable to those obtained by the best state of the art methods.

View PDFchevron_right
Evaluating the Effects of Textual Features on Authorship Attribution Accuracy
Reza Ramezani, M. Kahani

Abstract- Authorship attribution (AA) or author identification refers to the problem of identifying the author of an unseen text. From the machine learning point of view, AA can be viewed as a multiclass, single-label text-categorization task. This task is based on this assumption that the author of an unseen text can be discriminated by comparing some textual features extracted from that unseen text with those of texts with known authors. In this paper the effects of 29 different textual features on the accuracy of author identification on Persian corpora in 30 different scenarios are evaluated. Several classification algorithms have been used on corpora with 2, 5, 10, 20 and 40 different authors and a comparison is performed. The evaluation results show that the information about the used words and verbs are the most reliable criteria for AA tasks and also NLP based features are more reliable than BOW based features.

View PDFchevron_right
General Terms Authorship analysis
sara El

2014

The objective in this paper is to provide a review of the different studies done on authorship analysis. Focus is on outlining the Stylometric features that allow distinguishing between authors and on listing the diverse techniques used to classify an author’s texts.

View PDFchevron_right
Lexical-Syntactic and Graph-Based Features for Authorship Verification
Darnes Vilariño Ayala

In this paper we present the results obtained by an approach submitted to the author identification task of PAN 2013 which uses lexical, syntactic and graph-based features for constructing a representation model of document authors. In particular, the features extracted from the graph representation were obtained by means of the SubDue mining tool. As a classification model we have employed Support Vector Machines (SVM). The overall results have ranked our approach in the fifth place from around 17 teams.

View PDFchevron_right
Design and Implementation of a Machine Learning-Based Authorship Identification Model
Shabana Ramzan

Scientific Programming

In this paper, a novel approach is presented for authorship identification in English and Urdu text using the LDA model with n-grams texts of authors and cosine similarity. The proposed approach uses similarity metrics to identify various learned representations of stylometric features and uses them to identify the writing style of a particular author. The proposed LDA-based approach emphasizes instance-based and profile-based classifications of an author’s text. Here, LDA suitably handles high-dimensional and sparse data by allowing more expressive representation of text. The presented approach is an unsupervised computational methodology that can handle the heterogeneity of the dataset, diversity in writing, and the inherent ambiguity of the Urdu language. A large corpus has been used for performance testing of the presented approach. The results of experiments show superiority of the proposed approach over the state-of-the-art representations and other algorithms used for authors...

View PDFchevron_right
Examining multiple features for author profiling
Jose Palazzo Moreira De Oliveira, Edson Weren, Lucas Mizusaki

Authorship analysis aims at classifying texts based on the stylistic choices of their authors. The idea is to discover characteristics of the authors of the texts. This task has a growing importance in forensics, security, and marketing. In this work, we focus on discovering age and gender from blog authors. With this goal in mind, we analyzed a large number of features -ranging from Information Retrieval to Sentiment Analysis. This paper reports on the usefulness of these features. Experiments on a corpus of over 236K blogs show that a classifier using the features explored here have outperformed the state-of-the art. More importantly, the experiments show that the Information Retrieval features proposed in our work are the most discriminative and yield the best class predictions.

View PDFchevron_right
Automatic Author Profiling Based on Linguistic and Stylistic Features
Dipankar Das

The rapid expansion of blog and electronic data in Web 2.0 is abounding and thus it is becoming important to identify the author"s profile also. The problems of automatic identification of author"s gender and age based on linguistic and stylistic pattern have been a subject of increasingly research interest in the recent years. The research methodologies are also helpful for several other applications like criminal detection, security and author detection etc. We have used lexical, syntactic and structural features for identifying the gender and age group of the authors. We have employed the Decision tree classifier for classifying the author profile. We have achieved the accuracies of 56.83% and 28.95% for gender and age group classification, respectively.

View PDFchevron_right
Syntax-based features and Machine Learning techniques for Authorship Attribution
Kim Luyckx

2004

View PDFchevron_right
Automatic Author Profiling Based on Linguistic and Stylistic Features Notebook for PAN at CLEF 2013
Tanik Saikh

2013

The rapid expansion of blog and electronic data in Web 2.0 is abounding and thus it is becoming important to identify the author"s profile also. The problems of automatic identification of author"s gender and age based on linguistic and stylistic pattern have been a subject of increasingly research interest in the recent years. The research methodologies are also helpful for several other applications like criminal detection, security and author detection etc. We have used lexical, syntactic and structural features for identifying the gender and age group of the authors. We have employed the Decision tree classifier for classifying the author profile. We have achieved the accuracies of 56.83% and 28.95% for gender and age group classification, respectively.

View PDFchevron_right

References (78)

  1. Androutsopoulos, I., Koutsias, J., Chandrinos, K., Paliouras, G., & Spyropoulos, C. (2000). An evaluation of naive bayesian anti-spam fil- tering. In Proceedings of the Workshop on Machine Learning in the New Information Age (pp. 9-17). New York: ACM Press.
  2. Appelt, D., Hobbs, J., Bear, J., Israel, D., & Tyson, M. (1993). FASTUS: A finite-state processor for information extraction from real-world text. In Proceedings of the International Joint Conference on Artificial Intelligence (pp. 1172-1178).
  3. Argamon, S., Dodick, J., & Chase, P. (2005). The languages of science: A corpus-based study of experimental and historical science articles. In Proceedings of the 26th Annual Meeting of the Cognitive Science Society (pp. 157-162).
  4. Argamon, S., Koppel, M., Fine, J., & Shimony, A.R. (2003). Gender, genre, and writing style in formal written texts. Text, 23(3), 321-346.
  5. Argamon, S., & Levitan, S. (2005). Measuring the usefulness of function words for authorship attribution. In Proceedings of the 2005 ACH/ ALLC Conference. Retrieved from http://web.uvic.ca/hrd/achallc2005/ abstracts.htm
  6. Argamon, S., & Olsen, M. (2006, April). Toward meaningful computing. Communications of the ACM, 49(4), 33-35.
  7. Baayen, R.H., Halteren, H. van, & Tweedie, F. (1996). Outside the cave of shadows: Using syntactic annotation to enhance authorship attribution. Literary and Linguistic Computing, 7, 91-109.
  8. Belkin, N.J. (1993). Interaction with texts: Information retrieval as infor- mation-seeking behavior. Information Retrieval, 93, 55-66.
  9. Berry, M.J., & Linoff, G. (1997). Data mining techniques: For marketing, sales, and customer support. New York: Wiley.
  10. Burrows, J.F. (1987). Computation into criticism: A study of Jane Austen's novels and an experiment in method. Oxford, England: Clarendon Press.
  11. Butler, C.S. (2003). Structure and function-A guide to three major structural-functional theories (No. 63-64). Amsterdam: John Benjamins.
  12. Chaski, C.E. (1999). Linguistic authentication and reliability. In Proceed- ings of the National Conference on Science and the Law (pp. 97-148). San Diego, CA: National Institute of Justice.
  13. Chen, S.Y., Magoulas, G.D., & Dimakopoulos, D. (2005). A flexible inter- face design for web directories to accommodate different cognitive styles. Journal of the American Society for Information Science and Technology, 56(1), 70-83.
  14. Cowie, J., & Lehnert, W. (1996). Information extraction. Communications of the ACM, 39(1), 80-91.
  15. Cristea, D., Marcu, D., Ide, N., & Tablan, V. (1999). Discourse structure and co-reference: An empirical study. In D. Cristea, N. Ide, & D. Marcu (Eds.), The relation of discourse/dialogue structure and reference (pp. 46-53). New Brunswick, NJ: Association for Computational Linguistics.
  16. Cristianini, N., & Shawe-Taylor, J. (2000). An introduction to support vector machines. Cambridge: Cambridge University Press.
  17. de Vel, O. (2000). Mining e-mail authorship. In ACM International Confer- ence on Knowledge Discovery and Data Mining Workshop on Text Mining. Boston. Retrieved from http://www.cs.cmu.edu/~dunja/ WshKDD2000.html
  18. de Vel, O., Corney, M., Anderson, A., & Mohay, G. (2002). Language and gender author cohort analysis of e-mail for computer forensics. In Proceedings of the Digital Forensic Research Workshop. Syracuse, NY. (pp. 7-9).
  19. Fawcett, R.P., & Tucker, G.H. (1990). Demonstration of GENESYS: A very large, semantically based systemic functional grammar. In Proceedings of the 13th International Conference on Computational Linguistics (COLING-90) (pp. 47-49). Helsinki, Finland.
  20. Finn, A., Kushmerick, N., & Smyth, B. (2002). Genre classification and do- main transfer for information filtering. In F. Crestani, M. Girolami, & C.J. van Rijsbergen (Eds.), Proceedings of the 24th European Collo- quium on Information Retrieval Research. Glasgow, United Kingdom: Springer Verlag, Heidelberg, DE.
  21. Firth, J. (1968). A synopsis of linguistic theory 1930-1955.
  22. In F. Palmer (Ed.), Selected papers of J.R. Firth 1952-1959. London: Longman.
  23. Fritch, J.W., & Cromwell, R.L. (2001). Evaluating internet resources: Iden- tity, affiliation, and cognitive authority in a networked world. Journal of the American Society for Information Science and Technology, 52(6), 498-507.
  24. Grossman, D., & Frieder, O. (1998). Information retrieval: Algorithms and heuristics. Dordrecht, The Netherlands: Kluwer.
  25. Halliday, M.A.K. (1994). Introduction to functional grammar (2nd ed.). London: Arnold.
  26. Halliday, M.A.K., & Hasan, R. (1976). Cohesion in English. London: Longman.
  27. Holmes, D.I. (1998). The evolution of stylometry in humanities scholar- ship. Literary and Linguistic Computing, 13(3), 111-117.
  28. Hoover, D. (2002). Frequent word sequences and statistical stylistics. Liter- ary and Linguistic Computing, 17, 157-180.
  29. Kamps, J., Marx, M., Mokken, R.J., & Rijke, M. de. (2002). Words with attitude. In Proceedings of the 1st International Conference on Global WordNet. Mysore, India (pp. 332-341).
  30. Karlgren, J. (2000). Stylistic experiments for information retrieval. Unpub- lished doctoral dissertation, SICS.
  31. Kehagias, A., Petridis, V., Kaburlasos, V., & Fragkou, P. (2003). A compari- son of word-and sense-based text categorization using several classifica- tion algorithms. Journal of Intelligent Information Systems, 21(3), 227-247.
  32. Kessler, B., Nunberg, G., & Schütze, H. (1997). Automatic detection of text genre. In P.R. Cohen & W. Wahlster (Eds.), Proceedings of the 35th an- nual meeting of the Association for Computational Linguistics and 8th Conference of the European Chapter of the Association for Computa- tional Linguistics (pp. 32-38). Somerset, NJ: Association for Computa- tional Linguistics.
  33. Kjell, B., & Frieder, O. (1992). Visualization of literary style. In Proceed- ings of the IEEE International Conference on Systems, Man and Cyber- netics (pp. 656-661). Chicago: IEEE Press.
  34. Koppel, M., Akiva, N., & Dagan, I. (2003). A corpus-independent feature set for style-based text categorization. In Workshop on Computational Approaches to Style Analysis and Synthesis, 18th International Joint Conference on Artificial Intelligence.
  35. Kushmerick, N. (1999). Learning to remove internet advertisement. In O. Etzioni, J.P. Müller, & J.M. Bradshaw (Eds.), Proceedings of the 3rd International Conference on Autonomous Agents (agents'99) (pp. 175-181). Seattle, WA: ACM Press.
  36. Labov, W. (1973). Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press.
  37. Lewis, D., Schapire, R.E., Callan, J.P., & Papka, R. (1996). Training algo- rithms for linear text classifiers. In Proceedings of the 19th International Conference on Research and Development in Information Retrieval (pp. 298-306). New York: ACM Press.
  38. Marcu, D. (1997). The rhetorical parsing of natural language texts. In Meet- ing of the Association for Computational Linguistics (pp. 96-103). Morristown, NJ: ACL.
  39. Marcu, D. (1999). A decision-based approach to rhetorical parsing. In Pro- ceedings of the ACL'99 (pp. 365-372). Morristown, NJ: ACL.
  40. Martin, J.R., & White, P.R.R. (2005). The language of evaluation: Appraisal in English. London: Palgrave. http://www.grammatics.com/appraisal/
  41. Matthews, R.A.J., & Merriam, T.V.N. (1997). Distinguishing literary styles using neural networks. In E. Fiesler & R. Beale (Eds.), Handbook of neural computation (pp. 8). New York: IOP Publishing and Oxford Uni- versity Press.
  42. Matthiessen, C. (1983). Systemic grammar in computation: The nigel case. In Proceedings of the Meeting of the European Association for Computa- tional Linguistics (pp. 155-164). Morristown, NJ: ACL.
  43. Matthiessen, C. (1995). Lexico-grammatical cartography: English systems. Tokyo: International Language Sciences.
  44. Matthiessen, C., & Bateman, J.A. (1991). Text generation and systemic- functional linguistics: Experiences from English and Japanese. London, New York: Pinter, St. Martin's Press.
  45. McCallum, A., Freitag, D., & Pereira, F. (2000). Maximum entropy Markov models for information extraction and segmentation. In Pro- ceedings of the 17th International Conference on Machine Learning. Stanford, CA (pp. 591-598).
  46. McEnery, A., & Oakes, M. (2000). Authorship studies/textual statistics. In R. Dale, H. Moisl, & H. Somers (Eds.), Handbook of natural language processing (pp. 234-248). Philadelphia: Dekker.
  47. McKinney, V., Yoon, K., & Zahedi, F.M. (2002). The measurement of web- customer satisfaction: An expectation and disconfirmation approach. In- formation Systems Research, 13(3), 296-315.
  48. McMenamin, G. (2002). Forensic linguistics: Advances in forensic stylis- tics. Boca Raton, FL: CRC Press.
  49. Miller, G., Beckwith, R., Fellbaum, C., Gross, D., & Miller, K. (1990). Wordnet: An on-line lexical database. International Journal of Lexicogra- phy, 3(4), 235-312.
  50. Moore, J.D., & Pollack, M.E. (1992). A problem for RST: The need for multi- level discourse analysis. Computational Linguistics, 18(4), 537-544.
  51. Mosteller, F., & Wallace, D.L. (1964). Inference and disputed authorship: The federalist. Reading, MA: Addison-Wesley.
  52. Ng, V., & Cardie, C. (2002). Improving machine learning approaches to coref- erence resolution. In Proceedings of the 40th annual meeting of the Associ- ation for Computational Linguistics (pp. 104-111). Morristown, NJ: ACL.
  53. O'Donnell, M. (1993). Reducing complexity in a systemic parser. In Pro- ceedings of the 3rd International Workshop on Parsing Technologies. Tilburg, the Netherlands (pp. 10-13).
  54. Osgood, C.E., Succi, G.J., & Tannenbaum, P.H. (1957). The measurement of meaning. Urbana: University of Illinois Press.
  55. Pang, B., & Lee, L. (2004). A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts. In Proceed- ings of the 42nd ACL (pp. 271-278). Morristown, NJ: ACL.
  56. Pang, B., Lee, L., & Vaithyanathan, S. (2002). Thumbs up? Sentiment clas- sification using machine learning techniques. In Proceedings of the ACL-02 Conference on Empirical Methods in Natural Language Processing (pp. 79-86). Morristown, NJ: ACL.
  57. Patrick, J. (2004). The ScamSeek project: Text mining for finanical scams on the internet. In S. Simoff & G. Williams (Eds.), Proceedings of the 3rd Australasian Data Mining Conference (pp. 33-38).
  58. Platt, J. (1998). Fast training of support vector machines using sequential minimal optimization. In B. Scholkopf, C.J.C. Burges, & A.J. Smola (Eds.), Advances in Kernel Methods-Support vector learning. Cam- bridge, MA, MIT Press.
  59. Ponte, J.M., & Croft, W.B. (1998). A language modeling approach to infor- mation retrieval. In Proceedings of ACM SIGIR. New York: ACM Press.
  60. Roth, D., & Yih, W. (2001). Relational learning via propositional algorithms: An information extraction case study. In Proceedings of the International Joint Conference on Artificial Intelligence (pp. 1257-1263).
  61. Salton, G., & McGill, M. (1983). Introduction to modern information re- trieval. New York: McGraw-Hill.
  62. Schauer, H., & Hahn, U. (2001). Anaphoric cues for coherence relations. In G. Angelova, K. Bontcheva, R. Mitkov, N. Nicolov, & N. Nikolov (Eds.), Proceedings of the Euroconference Recent Advances in Natural Lan- guage Processing (RANLP-2001) (pp. 228-234). Tzigov, Bulgaria.
  63. Sebastiani, F. (2002). Machine learning in automated text categorization. ACM Computing Surveys, 34(1), 1-47.
  64. Shakespeare, W. (n.d.). The complete Moby Shakespeare. http://www- tech.mit.edu/Shakespeare/
  65. Stamatatos, E., Fakotakis, N., & Kokkinakis, G.K. (2000). Automatic text categorization in terms of genre, author. Computational Linguistics, 26(4), 471-495.
  66. Taboada, M., & Grieve, J. (2004). Analyzing appraisal automatically. In AAAI Spring Symposium on Exploring Attitude and Affect in Text. Menlo Park, CA: AAAI Press.
  67. Tang, R., Ng, K.B., Strzalkowski, T., & Kantor, P.B. (2003). Toward ma- chine understanding of information quality. In Proceedings of Annual Meeting of American Society for Information Science and Technology (Vol. 40, pp. 213-220).
  68. Teich, E. (1995). A proposal for dependency in systemic functional grammar-Metasemiosis in computational systemic functional linguis- tics. Unpublished doctoral dissertation, University of the Saarland and GMD/IPSI, Darmstadt, Germany.
  69. Torvik, V.I., Weeber, M., Swanson, D.R., & Smalheiser, N.R. (2005). A probabilistic similarity metric for Medline records: A model for author name disambiguation. Journal of the American Society for Information Science and Technology, 56(2), 140-158.
  70. Trudgill, P. (2001). Sociolinguistics: An introduction to language and society (4th ed.). New York: Penguin.
  71. Turney, P.D. (2002). Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews. In Proceedings of the 40th annual meeting of the ACL (pp. 417-424). Morristown, NJ: ACL.
  72. Tweedie, F., Singh, S., & Holmes, D. (1996). Neural network applications in stylometry: The Federalist Papers. Computers and the Humanities, 30(1), 1-10.
  73. Whitelaw, C., Garg, N., & Argamon, S. (2005, May). Using appraisal tax- onomies for sentiment analysis. In Proceedings of the 2nd Midwest Com- putational Linguistic Colloquium (MCLC 2005).
  74. Wiebe, J., McKeever, K., & Bruce, R. (1998). Mapping collocational prop- erties into machine learning features. In Proceedings of the 6th Workshop on Very Large Corpora (pp. 225-233). Morristown, NJ: ACL.
  75. Wiebe, J., Wilson, T., & Bell, M. (2001). Identifying collocations for recog- nizing opinions. In Proceedings of ACL/EACL 2001 Workshop on Collocation (pp. 24-31).
  76. Winograd, T. (1972). Understanding natural language. Orlando, FL: Academic Press.
  77. Witten, I.H., & Frank, E. (2000). Data mining: Practical machine learning tools with java implementations. San Francisco: Kaufmann.
  78. Yule, G.U. (1944). Statistical study of literary vocabulary. Cambridge, UK: Cambridge University Press.

Related papers

Systemic Functional Approach to Automated Authorship Analysis, A
Moshe Koppel

Most text analysis and retrieval work to date has focused on the topic of a text; that is, what it is about. However, a text also contains much useful information in its style, or how it is written. This includes information about its author, its purpose, feelings it is meant to evoke, and more. This article develops a new type of lexical feature for use in stylistic text classification, based on taxonomies of various semantic functions of certain choice words or phrases. We demonstrate the usefulness of such features for the stylistic text classification tasks of determining author identity and nationality, the gender of literary characters, a text's sentiment (positive/ negative evaluation), and the rhetorical character of scientific journal articles. We further show how the use of functional features aids in gaining insight about stylistic differences among different kinds of texts.

View PDFchevron_right
Stylistic text classification using functional lexical features
Casey Whitelaw

Journal of The American Society for Information Science and Technology, 2007

Most text analysis and retrieval work to date has focused on the topic of a text; that is, what it is about. However, a text also contains much useful information in its style, or how it is written. This includes information about its author, its purpose, feelings it is meant to evoke, and more. This article develops a new type of lexical feature for use in stylistic text classification, based on taxonomies of various semantic functions of certain choice words or phrases. We demonstrate the usefulness of such features for the stylistic text classification tasks of determining author identity and nationality, the gender of literary characters, a text's sentiment (positive/negative evaluation), and the rhetorical character of scientific journal articles. We further show how the use of functional features aids in gaining insight about stylistic differences among different kinds of texts.

View PDFchevron_right
Systemic Functional Features in Stylistic Text Classification
Casey Whitelaw

We propose that textual 'style' should be best defined as 'non-denotational meaning', i.e., those aspects of a text's meaning that are mostly independent of what the text refers to in the world. To make this more concrete, we describe a linguistically well-motivated framework for computational stylistic analysis based on Systemic Functional Linguistics. This theory views a text as a realisation of multiple overlapping choices within a network of related meanings, many of which relate to non-denotational (traditionally 'pragmatic') aspects such as cohesion or interpersonal distance. Variations between relative frequencies of options within these systems corresponds to stylistic variation. Though full parsing in SFL remains a difficult unsolved problem, we present a software architecture which allows for efficient modelling and extraction of SFL entities for use in stylistic analyses. In support, we show results for two applications: classifying texts as financial scams and classifying scientific articles by field.

View PDFchevron_right
A Survey on Authorship Analysis Tasks and Techniques
Ercan Canhasi

SEEU Review

Authorship Analysis (AA) is a natural language processing field that examines the previous works of writers to identify the author of a text based on its features. Studies in authorship analysis include authorship identification, authorship profiling, and authorship verification. Due to its relevance, to many applications in this field attention has been paid. It is widely used in the attribution of historical literature. Other applications include legal linguistics, criminal law, forensic investigations, and computer forensics. This paper aims to provide an overview of the work done and the techniques applied in the authorship analysis domain. The examination of recent developments in this field is the principal focus. Many different criteria can be used to define a writer’s style. This paper investigates stylometric features in different author-related tasks, including lexical, syntactic, semantic, structural, and content-specific ones. A lot of classification methods have been ap...

View PDFchevron_right
The Role of Linguistic Feature Categories in Authorship Verification
Hossam Ahmed

Authorship verification is a type of authorship analysis that addresses the following problem: given a set of documents known to be written by an author, and a document of doubtful attribution to that author, the task is to decide whether that document is truly written by that author. A combination of a similarity-based method and relevant linguistic features is used to achieve high accuracy authorship verification. The method is an author-profiling approach that dispenses with negative-evidence training data, and a number of lexical, morphological, and syntactic features and feature ensembles are used to determine optimal feature use. The method-feature combination is applied to a test corpus of 31 Classical Arabic books and substantially outperforms best available baselines (with 87.1% accuracy). The varying performance of different features and feature ensembles indicate that Classical Arabic authors are less free to individualize their style lexically or morphologically than when involving syntactic structures.

View PDFchevron_right

Related topics

  • Digital Humanities
  • Authorship Attribution
  • Systemic Functional Linguistics

    • Cited by

    The Rest of the Story: Finding Meaning in Stylistic Variation
    Moshe Koppel

    The Structure of Style, 2010

    As we have seen, there are many diverse concepts of what we mean when we speak of the 'style' of a work; we may speak, for example, of an individual's specific style of artistic expression , of style as a means of implicitly contextualizing a work as part of a certain genre , of style as a means of conveying emotion [26], or of style as a concept organizing how observers tend to interpret a work . In this chapter, we explore automated methods for stylistic analysis of written texts (computational stylistics), formulated as having roughly two research goals: (a) develop methods to automatically distinguish texts with a certain stylistic character from those of another, and (b) distill an interpretable representation of the difference between such stylistic characters. We seek an inclusive view of the nature of style, endeavoring to consider models and methods covering the individual style of a genius as well as the generic style of a collective, style's communicative functions as well as its social determinants, and the intentions of the author as well as the potential reaction of the reader.

    View PDFchevron_right
    Preliminary Studies on Register Function of Chinese Function Words
    Chu-Ren Huang

    Lingua sinica

    This paper explores the effectiveness of Chinese function words in register studies based on text mining and attempts to determine their register specific functions based on quantitative methods. Following the hypothesis that function words can be used as features to differentiate different registers, our text clustering experiment demonstrates that different registers in Chinese can in fact be classified in terms of distribution of function words. Our correspondence analysis shows that the correlations between function words and different registers vary, hence some function words have register-specific functions. In addition, the correlation coefficients between function words and various registers can determine the register-specific function of function words. Our quantitative model of register specific function of function words makes an important contribution to our understanding of the correlation between function words and registers.

    View PDFchevron_right
    Authorship Attribution by Functional Discriminant Analysis
    Chahrazed Kettaf

    2019

    Recognizing the author of a given text is a very difficult task that relies on several complicated and correlated criterias. For this purpose, several classification methods are used (neuronal network, discriminant analysis, SVM...). But a good representation of the text that keeps the maximum of the stylistic information is very important and has a considerable influence on the result. In this paper, we will tackle the problem of the authorship attribution for very long texts using the discriminant analysis extended to the functional case after presenting the texts as elements of a separable Hilbert space.

    View PDFchevron_right
    Language use reflects scientific methodology: A corpus-based study of peer-reviewed journal articles
    Shlomo Engelson Argamon

    Scientometrics, 2008

    Recently, philosophers of science have argued that the epistemological requirements of different scientific fields lead necessarily to differences in scientific method. In this paper, we examine possible variation in how language is used in peer-reviewed journal articles from various fields to see if features of such variation may help to elucidate and support claims of methodological variation among the sciences. We hypothesize that significant methodological differences will be reflected in related differences in scientists' language style. This paper reports a corpus-based study of peer-reviewed articles from twelve separate journals in six fields of experimental and historical sciences. Machine learning methods were applied to compare the discourse styles of articles in different fields, based on easily-extracted linguistic features of the text. Features included function word frequencies, as used often in computational stylistics, as well as lexical features based on systemic functional linguistics, which affords rich resources for comparative textual analysis. We found that indeed the style of writing in the historical sciences is readily distinguishable from that of the experimental sciences. Furthermore, the most significant linguistic features of these distinctive styles are directly related to the methodological differences posited by philosophers of science between historical and experimental sciences, lending empirical weight to their contentions.

    View PDFchevron_right
    Bootstrap Domain-Specific Sentiment Classifiers from Unlabeled Corpora
    Dell Zhang

    Transactions of the Association for Computational Linguistics, 2018

    There is often the need to perform sentiment classification in a particular domain where no labeled document is available. Although we could make use of a general-purpose off-the-shelf sentiment classifier or a pre-built one for a different domain, the effectiveness would be inferior. In this paper, we explore the possibility of building domain-specific sentiment classifiers with unlabeled documents only. Our investigation indicates that in the word embeddings learned from the unlabeled corpus of a given domain, the distributed word representations (vectors) for opposite sentiments form distinct clusters, though those clusters are not transferable across domains. Exploiting such a clustering structure, we are able to utilize machine learning algorithms to induce a quality domain-specific sentiment lexicon from just a few typical sentiment words (“seeds”). An important finding is that simple linear model based supervised learning algorithms (such as linear SVM) can actually work bett...

    View PDFchevron_right
    580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved