Kakuta Haruo---Decoding Japan---

My Photo
Name:
Location: Sakai, Osaka, Japan

Monday, July 15, 2013

5 Sentence Patterns?

How, or of what, is a sentence composed? In other words, what are the constituents of a sentence? At school, 5 Sentence Patterns with S = Subject, V = Verb, C = Complement, O = Object, and M = Modifier is dominant. However, in their experiment (J. L. Mclelland & A. H. Kawamoto, 'Mechanisms of Sentence Processing: Assigning Roles to Constituents of Sentences', "Parallel Distributed Processing Volume 2", MIT Press, Cambridge, 1986), Messrs. McClelland and Kawamoto employ a kind of cases such as A = Agent, V = Verb, P = Patient, I = Instrument, M = Modifier, F = (implied) Food, S = Self. They even proceed to argue the importance of "cross-connections among the case role units" and "back-connections from the case role units to the sentence-structure units". This matches my empirical knowledge: "Cross-connections among the case role units would add a number of other advantages, as well. They would allow competition among alternative interpretations of the same word at the case-frame level, so that the stronger of two competing interpretations of the same word could effectively suppress the weaker." (p.311) "the addition of back-connections from the case role units to the sentence-structure units. This, too, could have several beneficial effects on the performance of the model. In an extended version of the model with cross-connections and back-connections, the computation performed by the present version of the model would be just the first step in an iterative settling process. This settling process could be used to fill in the features of one reading or another of an ambiguous word at the sentence level, based on the emerging pattern at the case role level. Once filled in, these features could then add to the support for the 'dominant' reading over other, initially partially activated readings --- the whole network would, in effect, drive itself into a stable 'corner' that would tend to represent a coherent interpretation at both the sentence and the case role level." (p.311) "it will become crucial to train the sentence and case role units to represent just the needed conjunctions of features." (p.312)

Saturday, July 13, 2013

Visiting Universities as a part of Career Education

The first-year students of Senyo Senio High School formed groups of 20-80 and visited various universities last Thursday; Osaka University, Kobe University, Wakayama University, Nara University of Education, Mie University, Kobe City University of Foreign Studies, the University of Hyogo, the University of Shiga Prefecture, and the University of Tokushima. I led 40 students to Nara. Our bus entered the college gate and turned right to stop just in front of Yamada Hall (Students Hall). Its auditorium became our base of the day. After a short account of the university, we were divided into 2 groups of 20 students, and visited the University Library, Information Processing Facilities, and the Center for Educational Research. After the tour, we had lunch at the cafeteria. They were holding Oita Prefecture "Fair", and I resolutely had "tori-dukushi don" (an all chicken bowl of rice, topped with 2 kinds of chicken somethings and a boiled egg). A professor specializing in Special Needs Education gave us a rough sketch of the university in the auditorium, and, later, an alumnus of Senyo who studies at the university gave us a short introduction of his campus life and answered questions of our students, that is, his juniors. The bus left the woody campus sometime after 14:00 into the hot air of the town, and it took us a little bit more than an hour to get back to our school.

Tuesday, July 09, 2013

The Implementation Guidelines for the 3rd LL Class

(1) Start your browser. (2) Open the following web-page: http://www.shinko-keirin.co.jp/keirinkan/digital/ (3) Click the button under “Vision Quest English Expression I Advanced” to open: http://www.shinko-keirin.co.jp/keirinkan/digital/taiken/kou/eigo/visionquest_ad/index.html (4) Turn to page 8 of the electronic version textbook. (5) Open the file, “第3回LL授業.ppt”. (6) Let our students present “My Favorite Athlete” according to pages 1-2 of “第3回LL授業.ppt”. (7) Open page 3 of “第3回LL授業.ppt” and tell the students that they are going to introduce each other. (8) Tell the students to open their “Vision Quest English Expression I Advanced” to page 8, and show the page of its electronic version textbook on the screen with the browser. (9) Highlight the sentences of “STEP 1” on the page with a marker the page has, suggesting the students to use those sentences to interview each other. (10) Have the students practice reading aloud those sentences. (11) Highlight the column under “STEP 3” on the page with a marker the page has, suggesting the students to use the format to write introductions of their friends. (12) Open the window of “第3回LL授業.ppt” again to page 4. (13) Start a model interview between JET and ALT, and show a model introduction according to pages 4-6. (14) Turn to page 7, and have the students interview each other and write introductions.