added details of final project

grading.md

@@ -101,5 +101,5 @@ Your report should provide a complete description of what you did in your projec
 ### Writing advice:
 
 - **Use specific examples to bolster high-level conclusions**: for example, if you find that unigram text features predict retweets, then show some of the words which are the strongest predictors (like Tan et al); if you find network centrality correlates with microfinance participation, give some examples of highly-central individuals and non-central individuals, and describe their outcomes. *This is especially important if you use unsupervised techniques such as clustering*, since the output of these techniques is otherwise opaque to the reader.
-- Well-designed graphs and tables are more helpful than lots of text. Use the papers that we have read in class for guidance on how to write and structure your report, but keep in mind that each paper has its own idiosyncrasies, so think critically about what does and doesn't work in these papers.
-- Similar to Wikipedia, citations should be provided for all factual statements, such as "Twitter language is noisy and ungrammatical" and "neural networks are the most powerful classification algorithm", to pick two examples completely at random. If text is taken verbatim from another source, it should be presented in quotations, with a citation; if the text is paraphrased from another source, a citation should also be included. I'm not picky about citation formats, but be sure to include all authors names, the full title, the year, and an unambiguous description of where the paper was published. 
+- Use **graphs and tables** to illustrate the main quantitative findings of your replication and extension. Use the papers that we have read in class for guidance on how to write and structure your report, but keep in mind that each paper has its own idiosyncrasies, so think critically about what does and doesn't work in these papers.
+- Similar to Wikipedia, **citations should be provided for all factual statements**, such as "Twitter language is noisy and ungrammatical" and "neural networks are the most powerful classification algorithm", to pick two examples completely at random. If text is taken verbatim from another source, it should be presented in quotations, with a citation; if the text is paraphrased from another source, a citation should also be included. I'm not picky about citation formats, but be sure to include all authors names, the full title, the year, and an unambiguous description of where the paper was published.