Download FinalProjectInfo

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
CS311 Introduction to Artificial Intelligence – Fall 2016
Instructions for Final Project
For the final project, you will research and investigate an artificial intelligence topic of
your choice (with approval). You will work on this project in groups of 2-3 students (if
you prefer to work on your own, please come talk to me). I will provide some suggested
topics below but anything that is related to artificial intelligence is likely to be
appropriate.
The first stage of the project will consist of a brief literature review of your selected area.
Based on that, you will then select a more specific research problem and write a proposal
for your project. I will be available to help you with your proposal and progress
throughout the semester – do not hesitate to ask for help. In addition to the proposal, you
will be required to submit a progress report and final report. Due dates:



Proposal: Monday October 24th, by 11:15am (hardcopy in class).
Progress Report: Monday, November 21st by 11:15am (hardcopy in class).
Final Report: Wednesday December 14th by 10pm (via Moodle link).
More details on the project and possible topics are below:
Literature Review: You should first decide on one (or a few) general research areas
related to artificial intelligence that interest you. Once you decide on an area, research
some papers related to this area to identify a specific research problem that you would
like to investigate. For example, your general research area may be Machine Learning
and your specific research problem may be predicting Twitter tweets from politicians. In
class, we will quickly review how to search for research papers to help you with this step.
[7 pts] Proposal: The proposal should include the literature you reviewed to select your
research problem, a brief background of your topic, and your goals for the project. At this
stage your goals may be vague and broad, though if you have very specific goals in mind,
please include them in the proposal.
[15 pts] Progress Report: For the progress report, you are expected to specify your
goals much more clearly, typically in the form of a specific research problem you wish to
resolve. By this time you should have made some progress on your project so you should
also describe this progress (did you start implementing an algorithm/tool? Did you start
looking into data sources?) You should also outline your planned approach towards
satisfying your goals based on the progress you have made so far. Your report should
describe your plan to achieve your stated goals and any data you may use.
[15 pts] Presentation: During the last week of classes, you and your group will either (1)
give a 10-15 minute presentation about your project or (2) present your project results on
a webpage. Your project does not have to be complete by this time so you may present
your progress so far.
[63 pts] Final Report: Your final report will summarize your research and should
include at least the following: motivation, related work, background, the steps you took to
obtain your results, a discussion about your data and why you chose this dataset, your
results, and a conclusion. (Please see the Final Paper Template that I will hand out later in
the semester for more details). Remember that a negative result – i.e. discovering that a
hypothesis is false or that some methodology does not work is often an important result.
Below are some *possible* project ideas:







Create an intelligent player of a single or two-player game. Examples of games
include Tic-Tac-Toe, Mancala, Connect Four, 8-Queens, Minesweeper.
Examine different search algorithm performance on some tasks (for example, A* vs.
IDS) for route finding on specific maps.
Design a CSP solver for scheduling based on data from doodle polls.
Design, implement and evaluate a machine learning algorithm for
o text or spam filtering
o recommending news articles, movies, or music from the web based on a user’s
personal preferences
o a personal assistant for meeting scheduling
o classification of molecular sequences
Design a genetic algorithm to solve a computationally hard problem (for example,
related to cryptography)
Clustering algorithms for classification
Neural Networks for stock market prediction, image recognition.
If you would like some help coming up with a topic, please come talk to me!