Web and Network Data Science: Modeling Techniques in Predictive Analytics

C. Case Studies

This appendix reviews cases in web and network data science and modeling techniques in predictive analytics, as we have in the chapters of the book. Data for the cases are available in the public domain or are provided on the book’s website: http://www.ftpress.com/miller/.

C.1 E-Mail or Spam?

This case concerns the automatic detection of junk e-mail or spam. Spam is unsolicited and unwanted commercial e-mail, mass mailings such as advertisements for products, get-rich schemes, chain letters, and adult erotic literature. Spam is usually sent to people on mailing lists and newsgroups. This type of e-mail activity is considered unethical because the full cost of sending the messages is not borne by the senders and because recipients ...