Learning from Authoritative Security Experiment Results
Submission Guidelines: Background and Purpose
In his keynote speech at the 2012 LASER workshop, Dr. Roy Maxion of Carnegie Mellon University suggested a simple way to help improve the quality of experimental research for cyber security: to begin requiring that papers follow a specific structure, including a structured abstract that concisely and clearly summarizes the whole story of the work detailed in the paper.
As a result of round table discussion at the LASER workshop, the organizing committee decided to try this approach in the 2013 workshop. To that end, we are inviting structured abstracts for comment and review to help authors refine their abstracts prior to developing full papers. Workshop participants who do not have a paper in the workshop are also encouraged to take advantage of the abstract submission and review process to help them improve their abstract development skills. We will continue to accept and review abstracts after the official paper deadline. All abstracts deemed relevant by the PC will be available on the laser-workshop.org website before the conference, but they will not be part of the proceedings.
Researchers should not have to read a whole paper to determine what the research described in the paper is about. The idea behind a structured abstract is to avoid that problem by giving the reader a concise summary of the whole study and reducing his overall cognitive load. We also recommend using bold face as section headers in the abstract to make it easier on the reader. Finally, a well-written structured abstract will provide a good outline for an author to use in developing a full paper and for conducting meta analyses.
Structured Abstract Guidelines
Abstracts should be 200-500 words and less than one page in length. They should contain concise statements that tell the whole story of the study, presented in a consistent structure that facilitates quick assessment as to whether or not the paper may meet the reader's needs and warrant reading the full paper. Essential elements of structured abstracts are: background, aim, method, results, and conclusions.
- Background. State the background and context of the work described in the paper.
- Aim. State the research question, objective, or purpose of the work in the paper.
- Method. Briefly summarize the method used to conduct the research, including the subjects, procedure, data, and analytical method.
- Results. State the outcome of the research using measures appropriate for the study conducted. Results are essentially the numbers.
- Conclusions. State the lessons learned as a result of the study and recommendations for future work. The conclusions are the "so what" of the study.
By using this format for an abstract, the author has a good structure not only for his or her paper but also for creating slides to present the work.
Here is an example abstract from the below citation (140 words) of a LASER 2012 paper:
Kevin S. Killourhy and Roy A. Maxion. 2012. Free vs. transcribed text for keystroke-dynamics evaluations. In Proceedings of the 2012 Workshop on Learning from Authoritative Security Experiment Results (LASER '12). ACM, New York, NY, USA, 1-8.
- Background. One revolutionary application of keystroke dynamics is continuous reauthentication: confirming a typist’s identity during normal computer usage without interrupting the user.
- Aim. In laboratory evaluations, subjects are typically given transcription tasks rather than free composition (e.g., copy- ing rather than composing text), because transcription is easier for subjects. This work establishes whether free and transcribed text produce equivalent evaluation results.
- Method. Twenty subjects completed comparable transcription and free-composition tasks; two keystroke-dynamics classifiers were implemented; each classifier was evaluated using both the free-composition and transcription samples.
- Results. Transcription hold and keydown-keydown times are 2–3 milliseconds slower than free-text features; tests showed these effects to be significant. However, these effects did not significantly change evaluation results.
- Conclusions. The additional difficulty of collecting freely composed text from subjects seems unnecessary; researchers are encouraged to continue using transcription tasks.
Full papers should be 6-10 US letter sized pages (8.5"x11"), inclusive of tables, figures, and references. They should follow the latest ACM Proceedings Format (updated May 2013) and be submitted using Option 2, WITH the permission block. Papers must comply with the template margins and BE SUBMITTED IN PDF FORMAT.
Please include page numbers on submitted papers to aid reviewers. Page numbers should be excluded from final camera-ready papers. Additional guidelines for camera-ready papers will be provided to accepted paper authors.
Full papers should provide details sufficient that a reviewer can determine the validity of the experiment(s) conducted and repeat the experiment, if so desired. In addition to the title and author, suggested section headings are:
- Structured abstract (following the above guidelines)
- Problem being solved
- Background and related work
- Method (to include apparatus/instrumentation, materials, subjects/objects, instructions given to subjects, design, and procedure)
- Endnotes and footnotes
By using a predictable structure for content, the author is helping his readers because they know what to expect in each section. Further, it is easier for researchers to read sections in the order they choose and also to find something particular in the paper after reading. Whether the author chooses to follow this format or not, the information in the bolded sections is required.