Natural Language Processing

Student projects

Guidelines

  1. Develop a specific question about human behavior or cognition and address it using publicly available natural language data.
  2. Obtain natural language data from the resources listed below or other resources.
  3. Keep to the timeline.

Groups

Group Members
1 52 61 70 68 61 65 6c 20 45 64 69 73 6f 6e 20 4d 6f 73 74
54 61 6d 61 72 61 20 4c 6f 74 7a
50 65 6c 69 6e 20 53 69 6c 61 20 47 65 6c 6d 65 7a
52 6f 62 69 6e 20 42 72 c3 bc 67 67 65 6d 61 6e 6e
2 53 61 6d 75 65 6c 20 5a 65 69 73 65 72
44 61 6d 69 61 6e 20 50 61 72 6f
44 61 6d 69 61 6e 6f 20 4b 65 73 73 6c 65 72
4a 61 6e 20 42 69 74 74 65 72 6c 69
3 4e 69 65 76 65 73 20 53 63 68 77 61 62
4c 6f 72 69 73 20 4a 65 69 74 7a 69 6e 65 72
4d 79 72 61 20 4f 6c 69 76 69 61 20 46 69 73 63 68 65 72
4 44 61 6e 69 65 6c 65 20 55 72 73 6f
4d 61 72 69 65 20 41 6e 74 6f 6e 69 61 20 4a 61 6b 6f 62
55 64 61 79 20 47 c3 bc 7a 65 6c
4d 61 72 67 61 75 78 20 44 75 73 6f 75 6c 69 65 72

Project timeline

When What
29.09. Building groups
29.09. - 27.10. Find 2 project ideas and corresponding datasets and prepare project pitches
27.10. Pitch project proposals.
27.10. - 24.11. Project work and support meetings.
24.11. In-class markdown presentations.
24.04. - 15.05. Finalize project and presentation. Support meeting.
15.05. Deliver final presentation. Maximum 10 minutes.

Resources

Specific resources Books: Books, nGram Twitter: Various, twitteR Presidential speeches: CoPS Reviews: Movies Blogs: BAC Email: Enron News: Headlines Subtitles: Opensubtitles

Lists and search engines Google dataset search NLP datasets Kaggle

Support meetings

Each group must participate in least two 45 minute support meetings, one prior the markdown presentation and one prior the final presentation. Support meetings take place during the regular seminar slot, with two support meetings scheduled per slot. Support meetings will be scheduled on October 27 and November 24, respectively. If necessary, each group can request one additional 45 minute support meeting in one of two project work blocks. Given the limited time available the group should prepare a set of concrete questions, which should be submitted at least 24h prior to the support meeting.

Presentations

During the course, each group will give four presentations, two project pitches, one markdown presentation, and one final presentation. These presentations serve two important purposes. First, they will help you advance and track the progression of your project and invite helpful feedback from other seminar participants. Second, they provide valuable opportunities for practicing the communication of data analytic projects in various formats. The maxim for these presentations is to tell good stories that inform and entertain the audience. Some helpful advice can be found here.

Presention Constraints Description
Project pitch Max 3 minutes and 3 slides per project. Introduce and motivate a research questions and explain in very broad strokes how the chosen data set may help answer the question.
Markdown presentation 10 minutes, no slides, just a knitted markdown Briefly introduce and motivate question, then show, using your knitted markdown document, your progress in addressing the question and discuss outstanding issues.
Final presentation 10 minutes, max 7 slides Clean final presentation that presents a full, easy-to-follow story arc from a motivated question to conclusions derived from analytical results.