LINGUIST List 4.14

Wed 14 Jan 1993

Qs: Computational Linguistics, Quotation Source

Editor for this issue: <>

Directory

1. , Searching for implementations of the Disputed Authorship Index
2. , Source of quotation
3. Judith Klavans, computational linguistics
4. Penni Sibun, language discrimination

I'm searching for implementations of the Disputed Authorship Index, an
algorithm used in forensic linguistics that compares word usage in two
documents of similar length where one author is known and the other
unknown, generating a probability that the two documents were produced by
the same individual. I would also be interested in partial implementations
such as type/token overlap algorithms. I have access to part-of-speech
recognizers and some on-line dictionaries, though if a system demands a
particular resource set please let me know.

I work best in C or Pascal, but will happily accept code in any (uncompiled
:-) language. If there is interest, I will post a summary of responses.
Thanks in advance.

Best regards,

Curtis D. Frye
The MITRE Corporation, ATTN MS 267, 7525 Colshire Dr., McLean, VA 22102 USA
(703) 883-7843
cfryeciis.mitre.org (preferred) or cdfryeosprey.mitre.org or
cfryelinus.mitre.org or (last resort) cfrye%mdfmitre.org

This past fall, in teaching a course on Language and Women, I ran across
the following quote, for which I have been unable to find the citation. I
would be grateful to any party who can correspond with me directly on this.

The need to please marks women as subordinates, though often they are petted
subordinates, for pleasing is of course a delightful gift to receive.
Success at anything is enjoyable, and a woman who knows how to please others
may well be pleased with herself. There is, however, a problematic and
generally adverse effect within a woman's psyche when pleasing becomes a
political means to be used in any relationship at all, for any end: it is
mentally and morally confusing. ... One hesitates before one speaks
candidly
lest one seem too outspoken, for that may mean that important confidences will
not be offered in the future. Women's lives are molded by nuances like these.

Dear Colleagues,
I am putting together a new syllabus for a one-year course
in Computational Linguistics.
Although I taught a one-semester course to MA CS students
a few years ago, I understand there are new materials
out. Thus, I would like to gather opinions before writing
a syllabus. So I am asking for opinions, both from faculty,
and especially from students.

If you _teach_ CL:
1. what texts and materials have you used?
2. which have been good and which bad from the point of
view of your teaching, your ability to use the materials,
student complaints, etc.
3. what sorts of assignments have you given?

If you _took_ a course in CL:
1. same
2. same, from the point of view of what you learned
3. same, but let me know which kinds of assignments you
felt you learned the most from.

Since there will probably be a mixture of CS and Linguistics
students, I am particularly interested in knowing which
material you felt is optional/oblig for each type.

If you know of someone who has taught such a course, either
one-semester or two, and who is likely NOT to be on LINGUIST
(e.g. people from CS), I would appreciate it if you would
either forward this note to them, or let me know where to
find them.

Please answer _directly_ to me and I will summaraize for the
list. All syllabi are welcome, and will be acknowledged.

Thank you in advance,
Judith Klavans

i'm interested in hearing about any research on language
discrimination for written text. in particular, i'd like to know
about any methodologies for looking at a page of text and figuring out
what language it's in, using clues such as alphabet, distribution of
characters, and so forth.

please reply directly to me and i can summarize.

thank you very much.

 --penni sibun
 sibunparc.xerox.com