言語教育学演習2014

FrontPage

言語教育学演習2014（木３　前・後期） †

教室 †

• 201教室に変更になりました（5月15日より）

講義題目（和文） †

• コーパスを用いた学習者英語分析と教材作成への応用 (J)

授業の目標 †

• コーパス言語学の基礎的概念の習得とその応用分野としての CEFR 準拠の学習者データのコーパス分析の概要を知る。

授業の概要 †

• コーパス言語学の基礎概念をテキストを購読しながら学ぶと同時に，コーパス処理の基礎を習得する。　

授業の計画 †

• テキストを用いた専門知識の理解と Sketch Engine というコーパス検索システムの使用法に習熟する。

第1回(4/10)　イントロダクション：コーパスと言語教材作成
第2回(4/17)　コーパス言語学の基礎（１）：コーパスの定義と歴史的変遷　(CBLS Unit A1)
第3回(5/ 1)　コーパス言語学の基礎（２）：代表性・バランス・標本      (CBLS Unit A2) + 宿題（Unit A5・A8 --> summary in English)
第4回(5/15)　コーパス言語学の基礎（３）：Markup & annotation         (CBLS Unit A3　& A4) + 宿題（Unit A7 --> summary in English)
第5回(5/29)　コーパス言語学の基礎（４）：コーパスと応用分野          (CBLS Unit A10)
第6回(6/ 5)　コーパス言語学の基礎（５）：コーパスと辞書              (CBLS Unit C1) + BNCWeb の基本的な使い方とコロケーション統計
第7回(6/12)　コーパス言語学の基礎（６）：コーパスと文法研究          (CBLS Unit C2) + Sketch Engine を使った help (to) の検索実習
第8回(6/19)　コーパス言語学の基礎（７）：コーパスと言語習得研究      (CBLS Unit C3)
第9回(6/26)　コーパス言語学の基礎（８）：コーパスと翻訳研究          (CBLS Unit C6)
第10回(7/ 3)　CEFR準拠の学習者データの分析（１）：CEFR の概要と English Profile
第11回(7/10)　CEFR準拠の学習者データの分析（２）：English Profile の学習者データ分析 の概要
第12回　CEFR準拠の学習者データの分析（３）：エラータグ付与の実際
第13回　CEFR準拠の学習者データの分析（４）：基準特性抽出の研究概観
第14回　CEFR準拠の学習者データの分析（５）：学習者コーパスの種類と基礎的な処理 演習
第15回　まとめと検索処理のテスト

テキスト †

PDF †

テキストがくるまで以下をダウンロードして利用して下さい：

• Chapter A1
• Chapter A2
• Chapter A3
• Chapter A4

READING: †

Discussion questions †

• Chapter 1:

• What is a corpus? Discuss some common features by comparing different definitions.
• Why use computers to study language? What is your intuitive answer to this? What other reasons did you find in the text?
• Discuss the use of corpora and the use of intuition. Are they mutually exclusive?
• Is corpus linguistics a methodology or a theory?
• How different are corpus-based vs. corpus-driven approaches? Can you think of any concrete examples?

• Chapter 2:

• 2.2

• What is "representativeness"?
• What does it mean when Biber says "Representativeness refers to the extent to which a sample includes the full range of variabilityin a population." (p.13)
• What are "internal" and "external" criteria used to select texts for a corpus? (p.14)
• The authors say that it is problematical to use internal criteria as the primary parameters for the selection of corpus data. Why? (p.14)
• Explain what Biber calls a 'cyclical fashion'? (p.14)
• Static sample corpora, if resampled, may also allow the study of language change over time. (p.15) How?

• 2.3

• What are "general" vs. "specialized" corpora? How is representativeness achieved in these corpora?

• 2.4

• How is the acceptable balance of a corpus determined?
• Any claim of corpus balance is largely an act of faith. (p.16) What does this mean?
• Explain the design of the British National Corpus, using the terms 'domain', 'time', 'medium', 'demographic' and 'context-governed'. How is it balanced?
• Elaborate on the following statements:
  • Representativeness links to research questions. (p.18)
  • Representativeness is a fluid concept. (p.18)

• 2.5

• Explain the notion of sampling using the following terms:
  • sample/ population/ sampling unit/ sampling frame
• What is the difference between 'simple random sampling' and 'stratified random sampling'?
• Describe pros and cons of 'full text samples'

• Chapter 3

• 3.2
  • What are the three reasons for corpus mark-up? Discuss each case with complete examples.

• 3.3
  • Here, you should at least familiarize yourself with the following schemes:
    • COCOA (dated)
    • TEI (current standard) << website >>

         --> header vs. body
      Q1. What does the TEI header specify?
      Q2. What kind of information is in the TEI body?

• Corpus Encoding Standard (CES) & XCES << website >>

• 3.4
  • Please read the following webpage for your reference:
    • Introduction to character encoding

• Chapter 4

• What is corpus annotation and how is it different from corpus mark-up?

• 4.2
  • What are the four advantages for corpus annotation?
  • What are some of the criticisms against corpus annotation? What is the authors' response?

• 4.3
• Look at concrete examples for each type of annotation:

• POS tagging
  • Online tagging system by University of Illinois, Urbana Champaign

• Lemmatization
  • Online stemmer and lemmatizer (Python NLTK)

• Parsing
  • Online parser (Stanford)

• Semantic annotation
  • Lancaster USAS tag

• Coreference annotation
  • Image of coreference annotation

• Pragmatic annotation
  • Examples (MICASE pragmatic tags)

• Stylistic annotation
  • Example: Speech, Thought & Presentation

• Error tagging
  • Example: Granger (2003)

• Problem-oriented annotation

• Chapter 5-9

• Make a summary on your own

• Chapter 10

• Summarize the use of corpus data in the following areas briefly

• The major areas of linguistics
  • lexicographic and lexical studies (10.2)
  • grammatical studies (10.3)
  • register variation and genre analysis (10.4)
  • dialect distinction and language variety (10.5)
  • contrastive and translation studies (10.6)
  • diachronic study and language change (10.7)
  • language learning and teaching (10.8)

• Other areas which have started to use corpus data
  • Semantics (10.9)
  • Pragmatics (10.10)
  • Sociolinguistics (10.11)
  • Discourse analysis (10.12)
  • Stylistics and literary studies (10.13)
  • Forensic linguistics (10.14)

• What is the limitation of corpus data? (10.15)

Sketch Engine CQL memo †

• help + bare infinitives
  • Brown Family の CLAWS7 tagset の検索式：
    • [lemma="help"] [tag="VVI|VV0"]

• help + to + V
  • [lemma="help"] [lemma="to"] [tag="VVI"]

• make + a + adj(optional) + NOUN
  • [lemma="make"] [word="a"] [tag="JJ"]? [tag="NN1"]

カイ2乗検定など †

• js-STAR
  • web で計算できる統計

Sample data for error tagging †

• Raw file Download
• Tagged file Download

TUTORIALS: Corpus query tools †

• Sketch Engine

• 学内では IP auth をクリックすれば，どこからでもアクセスできます

• Antconc (for Windows & Mac)
  • Download

• Casualconc (for Mac)
  • Download

Concordance †

• Tutorial 資料(6/12)

• Tutorial 資料(5/29)

• Tutorial 資料(5/1)

• Tutorial 資料(4/17)

• 練習問題(5/1)