The average number of graduation projects which will be guided by a faculty is 4. For the students who want to work in natural language processing, I suggest to apply as soon as possible at the beginning of the fall semester. I usually encourage my students to use Java as the programming language and to attend the course BLG 505- Natural Language Processing.
Natural Language Processing deals with the statistical and/or rule-based modeling of natural language from a computational perspective. It is an interdisciplinary field also referred as Computer speech and language processing, Human language technology, Computational linguistics. The goal of this new field is to get computers to perform useful tasks involving human language, tasks like enabling human-machine communication, improving human-human communication, or simply doing useful processing of text or speech. There are many application areas where we can use nlp methods. Some of them may be listed as as the following:
The topics that I want to guide this year may be one of the below ones:
- Making Morphological Disambiguation using Word Sense Disambiguation Techniques
- Preparation of a Multiword Extraction Software for some predetermined classes such as: Named Entities, duplications, dates, compound prepositions
- The investigation of training set size improvement by using artificial techniques in the performance of dependency parsing.
- Creation of new parsing algorithms for Turkish.
- Any idea for the following contest www.ericssonapplicationawards.com
http://www.facebook.com/Ericsson.Application.Awards
PS: I didn't have time for now to explain the details of each topic. The students are welcomed to discuss them in depth.
How do I work with my students?
- Meetings: I expect that my students visit me every week and report their on going work. It is totaly the student's responsibility to arrange the meeting times.
- Meetings: I expect you to send me a short report (as a checklist which consists of the items that we have discussed in our last meeting) after and before each meeting (by email). The email that you will send before each meeting will summarize the accomplished work (by checked items) that we have scheduled for the previous week. You should expect a read-receipt after each email that you send me in order to verify if I get it. The subject of your email should be in a specific format:
- Format for graduate stıdents: YLTEZ_StudentName_ReportNo_Type . The report type will be either TS (aftermeeting) or TO (before meeting) (example: YLTEZ_GülşenEryiğit_1_TS)
- Format for undergradute students: BOTEZ_StudentName_ReportNo_Type
- Writing your thesis: It's too late to write your thesis after everything is accomplished. I expect you to write regulary while you are making your research and implementing your system.
- The student will carry the entire responsibility of the consequences that will occur because of him/her (missed meetings, falling behind the scheduled work ... )
- The student should reserve enough time for me to read the material that he/she sends.