1. What data do Hare et al use?

Hare et al use 3-D topographical images of their sample bullets from the Hamby study. Each bullet is scanned six times, once per land, in order to gather the data. Hare et al use bulletr and x3pr packages in R to extract and organize their data into x3p files.

  1. In what ways do the methods used by Hare et al differ from the “traditional” methods of bullet matching?

Traditional methods use a comparison microscope and an examiner to determine whether two bullets are matches or not (matches being fired through the same barrel). Typically 6 consecutively matching striae is considered to be the threshold in declare two bullets a match. Hare et al take many features into account in order to declare a match or non-match. These features are used to train a Random Forest model which determines whether bullets are a match or non-match based on many decision trees.

  1. How do Hare et al use clustering to help perform bullet matching tasks?

In the video Hare et al do not specifically say that they used clustering as a method to match their bullets. They do however use the Random Forest package in order to separate the bullets between matches and non-matches based on similarities and differences in features of the bullets. Two bullets with similar features would likely be categorized as matches. In the same manner, bullets with much different features would be categorized as non-matches.

  1. Identify one statistics and/or probability concept in the presentation that you have not heard of before. Do a little bit of research (Wikipedia is ok) and try to describe it to someone who doesn’t know about it. You don’t need to read it in its entirety, but you should also consult this paper to see if there is more detail on your chosen topic than is presented in the webinar.

Cross-correlation is a measure of similarities between two separate samples. A value is given to quantify the similarities with values close to 1 and -1 being linear and values close to 0 being scattered. Cross-correlation is used to assess how similar two bullets are to one another in order to determine if they are a match or non-match. Pattern recognition is the primary focus of cross-correlation in how it relates to this study.