WEKO3
アイテム
授業分析におけるデータサイエンス活用の可能性 : 潜在的意味解析を用いた逐語記録の分節わけの試み
https://doi.org/10.18999/nueduca.66.1.1
https://doi.org/10.18999/nueduca.66.1.117d52773-4b50-4873-934b-4be21592a6b5
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
02_nuedu660101.pdf (3.0 MB)
|
|
| Item type | 紀要論文 / Departmental Bulletin Paper(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2020-04-01 | |||||
| タイトル | ||||||
| タイトル | 授業分析におけるデータサイエンス活用の可能性 : 潜在的意味解析を用いた逐語記録の分節わけの試み | |||||
| 言語 | ja | |||||
| その他のタイトル | ||||||
| その他のタイトル | The Possibility of Utilizing Data Science in Lesson Analysis : Attempts with the Segment Division of Transcripts using Latent Semantic Analysis | |||||
| 言語 | en | |||||
| 著者 |
坂本, 將暢
× 坂本, 將暢× SAKAMOTO, Masanobu |
|||||
| アクセス権 | ||||||
| アクセス権 | open access | |||||
| アクセス権URI | http://purl.org/coar/access_right/c_abf2 | |||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | The purpose of this study is to explore whether it is possible to apply data science to Lesson Analysis, expanding current knowledge in the field. In particular, an attempt at a segment division using Latent Semantic Analysis, one method of text mining, is the purpose of this study. The Lesson Analysis that the author is currently performing is one of other methodologies initiated by Shigematsu and Ueda in the 1950s. It has continued to be utilized in the Method of Education at Nagoya University for approximately 70 years. It requires an analysis using a transcript that potentially reveals the child's thought and learning processes. In this study the appearance of words contained in a transcript, and the subsequent performance of a segment division is the focus and first task of Lesson Analysis. In order to understand the lesson structure and consider each part of the structure, a lesson is divided into “paragraphs.” This is referred to as a segment division. Because there is no specific “solution for the segment division,” it is essential to explore a “reasonable parting” by reading the transcript deeply and perhaps numerous times, as well as discussing it passionately. The creation of such a lesson plan is not an easy task. Of course, whether a single segment is appropriate or not is questionable. For example, beginners without sufficient analysis skills may cease to make individual investigations toward successful discussions, resulting in an inactive group session, which does little to promote their studies. Thus, this research attempts to ascertain whether beginners can make segment divisions effectively by using data science. As a first step in this research subject, the author attempts to see whether valid clues can be obtained in a segment division using Latent Semantic Analysis. Data used in this study is in the form of a transcript of a lesson of social studies. The lesson has 123 remarks. The remark in the number is about half of the number of classes. Since the student is talking about their own opinion, the number of characters in one speech is often compared to other lessons. The reason the author chose this particular transcript is that it contains various topics in the remarks, and because there are many remarks, even if there is only one remark, the number of remarks is less. Analysis procedure performs a singular value decomposition using statistical software that calculates the strengths of the degree of association of utterances. The value is plotted in a twodimensional plane, with segments divided from the distance farthest from the origin. The analysis selected in this study was a characteristic 32 words. Also, in order to compare and verify this analysis, the author compared the results analyzed by 20 words and 10 words being the most frequent, with the results analyzed by setting the three-dimensional dimensions of compressing. As a result, some of differences in the distance from the origin, related to the clue of the segment division, generally found a result resembling the settings. As far as transcript used in this study, the Lesson Analysis is considered to be set words, number of words, or number of dimensions a, utilized relatively freely. | |||||
| 言語 | en | |||||
| 出版者 | ||||||
| 出版者 | 名古屋大学大学院教育発達科学研究科 | |||||
| 言語 | ja | |||||
| 言語 | ||||||
| 言語 | jpn | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||
| 資源タイプ | departmental bulletin paper | |||||
| 出版タイプ | ||||||
| 出版タイプ | VoR | |||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||
| ID登録 | ||||||
| ID登録 | 10.18999/nueduca.66.1.1 | |||||
| ID登録タイプ | JaLC | |||||
| ISSN（print） | ||||||
| 収録物識別子タイプ | PISSN | |||||
| 収録物識別子 | 1346-0307 | |||||
| 書誌情報 |
ja : 名古屋大学大学院教育発達科学研究科紀要. 教育科学 巻 66, 号 1, p. 1-12, 発行日 2019-10-01 |
|||||
| 著者版フラグ | ||||||
| 値 | publisher | |||||
