STYLISTIC VARIATION AS A BASIS FOR GENRE-BASED TEXT CLASSIFICATION

1
STYLISTIC VARIATION AS A BASIS FOR GENRE-BASED
TEXT CLASSIFICATION

• S. Sameen Fatima
• Dept. of Computer Science Engineering
• Osmania University
• Hyderabad
• (sameenf_at_hotmail.com)

2
BACKGROUND

• Q. What is classification (of text)?
• A. Classification is an important IR task in
  which one or more category labels are assigned to
  a document.
• Approaches to Classification (of text)
• Earlier approaches to text classification
  assigned labels to documents based on CONTENTS
• 1. Word-based techniques
• -statistical (tf,idf)
• - term/keyword searches
• Advantage Simple and can be automated
• Disadvantage Phrases cannot be extracted
• 2. Phrase-based techniques
• a) In-depth NLP Here we aspire to represent all
  the information in a text using context.
• -syntax
• -semantics
• -statistics
• Advantage General task-independent
  representation
• Disadvantage Costly, Not possible in
  polynomial time
• b) Information Extraction Here we delimit in
  advance, as part of the specification of a task,
  the semantic range of the output, the relations
  we will represent and other allowable fillers in
  each slot.
• Advantage It works well for a specific corpus

3
Limitations of the Earlier Approaches to Text
Classification

• Texts have, besides content, STYLE which has not
  been accounted for.
• It is the focus of this talk to present STYLE as
  a new basis for text classification

4
COMPUTATIONAL STYLISTICS

• The study of style or in other words the
  detection of patterns common to a writing is
  known as STYLISTICS.
• If stylistic analysis uses computer-aided methods
  and statistical methods for analysis of texts,
  the field of study is called COMPUTATIONAL
  STYLISTICS.

5
Related Work in Computational Stylistics

• 1. Pre-WWW Era
• - Author Attribution Studies
• Popular Mosteller and Wallaces study of
  anonymous essays published in THE FEDERALIST to
  identify the authors (Hamilton and Madison).
• Stylistic parameters sentence-length, content
  words(nouns, adjectives, verbs), function
  words(preposition, conjunction), use of by, from,
  and to, ..
• Came up with interesting result that content
  words were too subject-dependent and were not
  good discriminators, while function words were
  good discriminators.
• - Automatic Abstracting
• Borko and Chatman advanced the view that it
  seems possible to make stylistic distinctions
  between informative (discusses research)
  abstracts and indicative (discusses the article
  whichh descsribes the research) abstracts, based
  on form, voice, tense, focus of the abstract.
• - Teaching writing styles for different types of
  documents.
• Writers WorkBench program on ATT Unix.

6
Related Work in Computational Stylistics(contd)

• 2. WWW-Era (on-going)
• -Stylistic variation between the different genres
  found in the Wall Street Journal. (Jussi
  Karlgren, Troy Strazheim)
• Example Articles, Business News with
  tables, Business News, Lists of briefs,
  Editorials, letters, Briefs, Whats New,
  Tables.
• Use simple stylistic parameters
  characters/word, digits/keywords, words/sentence.
• - Establishing a genre palette for internet
  material.
• (Jussi Karlgren, John Dewe, Ivan Bretan)

7

• Definition of a Genre/Functional Style
• A set of documents with a perceived consistent
  tendency to make the same stylistic choices,
  specifically if it has an established
  communication functions, a functional style.
• Genres can have differing usefulness
• Genres in my work (Corpus)
• Editorials from Hindu
• Editorials from Hindustan Times
• Editorials from Times of India

8

• Hypothesis
• Editorials from each newspaper show a systematic
  and consistent difference in the choice of a
  presentation style, specifically to establish
  some intended communication function (aggressive,
  conservative, liberal)
• Aim of the Experiment
• To find a descriptive and predictive algorithm
  for classifying editorials from different
  newspapers based on stylistic features.

9
Mathematical Model

• Two models were explored to find which was
  applicable.
• 1. Vector Space Model - Used by Salton in the
  SMART system (IRS)
• 2. Euclidean Space Model.
• Euclidean Space Model
• An n-dimensional Euclidean space, En is defined
  as the set of all n-tules of real numbers (x1,
  x2, ., xn) where the Euclidean distance in En
  between 2 points x (x1, x2, ., xn) and y
  (y1, y2, ., yn) is defined by
• d(x,y) sqrt((x1-y1)2 (x2-y2)2 .
  (xn-yn)2)
• In our project Euclidean Space represents a
  Stylistic Space

10

• In the Vector Space Model distance between two
  points x and y is related by the angle ?(x,y)
  formed by the lines from each of the points to
  the origin, which is given by
• cos ?(x,y) (x . y) / ( (x .x)0.5 (y . y)0.5)
• This failed in stylistic analysis

11
Stylistic Profiling

• A method of identifying the stylistic features in
  the writing style of an individual or a group of
  people and to present them in a systematic way.
• 1. Lexical Features
• Percentage of interrogative pronouns
• Percentage of emphatic pronouns
• Percentage of prepositions
• Percentage of conjunctions
• Percentage of articles
• Percentage of action words
• Percentage of unique words
• 2. Structural Features
• Average words/sentence
• maximum sentence length
• Total no. of sentences
• Total no. of words
• Total no. of characters
• 3. Affective Features
• Percentage of passive sentences
• Flesch Reading Ease
• Coleman Liau Grade level

12
Classification Algorithm
1. Training Phase
90 FSPs
Training set consisting of 30 editorials each
from H, HT, TI
90 SPs
Conduct ANOVA test extract the SIGNIFICANT
FEATURES
Compute the mean for each of the significant
features for each newspaper
3 Prototypes
Feature Extraction (Lexical, Structural,
Affective)
P-H P-HT P-TI
2. Classification Phase
Least d(I,P-H), Classify as Hindu
FSP, I
Significant Feature Extraction
Compute the distance between I and each of the
prototypes from the training phase
New instance of editorial
Least d(I,P-HT), Classify as HT
Least d(I,P-TI), Classify as TI
13
Results

• 1. Data Collection (SP)
• 2. Results of identifying significant features in
  the training phase (FSP)
• One-tailed ANOVA test was carried out
• Null hypothesis No difference between the
  means
• Alternate hypothesis Means are different
• ratio of the variance estimates is
  calculated, FSb2/Sw2
• Sb2 Sw2 (Check for null hypothesis)
• Sb2 gt Sw2 (Check for alternate hypothesis)
• F gt Fcrit for a particular significance level,
  then we say that the means of the feature are
  significantly different
• 3. Results of the classification phase

14

• Performance Evaluation
• Following measures were computed
• Precision Number-classified-correctly/Number-tot
  al-classified
• Recall Number-classified-correctly/Number-releva
  nt-for-classification
• Conclusion
• The results of the experiment were positive. It
  was possible to classify editorials with a good
  degree of recall and precision

15

• Scope for further work
• Currently, it is not clear whether topic and
  style are two independent dimensions of variation
  in text, or they go hand in hand. This can be
  further explored by subclassifying editorials
  based on topic and then studying each of them for
  stylistic variations
• Applications
• - For classifying documents on the Internet based
  on GENRE
• - Relating FSPs of editorials to the reader
  profiles for each newspaper so as to establish
  any interesting relationship.