Write a presentation demonstrating the use of the Natural Language Processing techniques you’ve learned in the course. You will choose a data set, and use the tools of your choice to analyze the data. In your presentation, you will describe the data and present the results of your analysis in the categories shown below. You may use screen shots, diagrams, tables, and text in your slides. You will present this project during the last week of class. This project accounts for approximately one third of your grade in this class, so be sure to do the best work you can do.
Data:
You may choose any data set you would like to work on, as long as it contains at least 1000 distinct unstructured texts. This can be a collection of Twitter data, blog posts, e-mails, news reports, or similar data. The following links contain some example data sets:
Enron Email dataset: https://www.cs.cmu.edu/~./enron/ (Links to an external site.)
Stanford Sentiment Analysis dataset (Twitter): http://nlp.stanford.edu/sentiment/ (Links to an external site.)
FBS opinion mining data sets: http://www.cs.uic.edu/~liub/FBS/sentiment-analysis.html#datasets (Links to an external site.)
Cornell movie review data sets: https://www.cs.cornell.edu/people/pabo/movie-review-data/ (Links to an external site.)
Processing:
Once you’ve chosen a dataset, you will perform the following tasks:
Clean the data for processing
Create a corpus appropriate for your tool
Tokenize the data
Perform basic descriptive analysis of the data
Perform your chosen analytic
Test and evaluate your analytic
Reporting:
Your report should consist of the following items, supported by appropriate text, diagrams, and screen shots:
A description of the dataset, including where it was acquired from, the format of the data, and whether or not the data included any tagging.
Basic statistics about the data
Number of documents
Average length of document (characters and words)
Frequency distribution/lexical dispersion
Explanation of your analytic tool and method, including any source code you produce
Explanation of your training and test set creation
Results of testing your analytic
Accuracy
Precision
Recall
F-score
Performance (average processing time per document)
Findings of the analytic
Your conclusions, including anything notable you learned in the process of completing the project.
Deliverable
Your final report should consist of at least 15-20 slides of content. Remember to properly document your sources.
Last Completed Projects
| topic title | academic level | Writer | delivered |
|---|