Software Engineering
Project 2 - Research Paper Preparation (and a start on the KCP&L Undergraduate Competition)
15 Points
Due: Oct. 3, 2013 at 11:59:59pm

Introduction:

Working professionals are often asked to prepare a short research paper. Research papers can serve different purposes, but one common use it to provide a comparison between different technologies, techniques, or policies as well as a recommendation.

Objectives:

• Develop a research paper, which can be an important part of domain analysis, or just a useful tool for making decisions about technologies, techniques, and policies.
• Practice formal business writing.
• Get a cash prize and an A!

Assignment Structure:

The research paper will be split into four phases. You will complete two of these phases for Project 2:

1. Preparation: (Must be shown to the instructor by Sept. 27 and submitted on Oct. 3, 7.5 points)

• Select what you feel is an appropriate topic and clear it with your instructor.
  You can select a topic of interest to you that is relevant to Software Engineering and can be used to develop an unbiased comparison between at least one alternative. Unbiased means you may explain factual advantages and disadvantages in your research paper.  It is reasonable to use some personal opinion when there is some controversy about a topic.  For example, some literature may show conflicting information.  You should not use personal opinion of whether you like or dislike something in your research paper aside from describing why you chose to investigate a particular topic in your introduction.  For example, you may feel that the Agile process is a better methodology than Waterfall or vice versa.  This research paper is not an appropriate place to state an opinion such as "Agile is better than Waterfall," and proceed to say many good things about the Agile method using a successful Agile example and many bad things about Waterfall by using a failed Waterfall example. Instead, if you were comparing methodologies it would be appropriate to indicate features of each, using comparable examples, showing the utility of such features in an unbiased manner, and make a comparison that shows the advantages and disadvantages of each without showing personal bias toward either method.  There will be a 25% penalty for excessive personal bias in a research paper - this means if your paper is written like a news article and not a research paper, you will be penalized.  Example topics (just a few, you do not have to choose from this list) and some starting references include:
  • Software development methodologies
  • Waterfall - Royce, Winston (1970), "Managing the Development of Large Software Systems"
  • Spiral - Boehm, Barry (1986), "A Spiral Model of Software Development and Enhancement"
    
  • Dual Vee - Forsberg, Kevin and Mooz, Harold (1991), "The Relationship of System Engineering to the Project Cycle"
    
  • SCRUM - Schwaber, Ken (1995), "SCRUM Development Process"
  • Iterative and Incremental Development - DOD Standard 2167 (1985)
    
  • Novel programming paradigms or tools
  • Hadoop and MapReduce - Dean, Jeffery and Ghemawat, Sanjay (2004), " MapReduce: simplified data processing on large clusters"
    
  • HBase, MongoDB, Redis, Cassandra and NoSQL Databases - Lith, Adam and Mattson, Jakob (2010), "Investigating storage solutions for large data: A comparison of well performing and scalable data storage solutions for real time extraction and batch insertion of data"
  • Cray's Chapel Programming Language - Bradford L. Chamberlain , David Callahan , Hans P. Zima, "Parallel Programmability and the Chapel Language"
  • Accelerators, CUDA programming, OpenACC, and OpenCL
    
  • Modeling and design methodologies
  • Modeling Languages such as BPMN, EXPRESS, IDEF, UML/OCL, etc.
  • Design Patterns - Baroni, Aline (2003), "Design Patterns Formalization"
    
  • Architecture Patterns - Microsoft Application Architecture Guide, 2nd Edition - October 2009, Chapter 3: Architectural Styles and Patterns
  • Symbolic methods
  • Symbolic Execution for Software Testing
  • Testing
  • White Box - White Box Testing
    
  • Black Box - BCS SIGINT Standard for Component Testing
    
  • Gray Box - Achieving Both Model and Code Coverage with Automated Gray-Box Testing
  • Regression Testing -  Hiralal Agrawal, Joseph R. Horgan, Edward W. Krauser, Saul A. London (1993), "Incremental Regression Testing"
    
  • User Acceptance Testing
    
  • Concurrency, Parallelism, and Performance
  • Load Balancing - Willebeek-LeMair, Marc and Reeves, Anthony (1993), "Strategies for dynamic load balancing on highly parallel computers"
  • OpenMP, MPI, POSIX Threads, etc.
  • Parallel Debugging - Charles E. Mcdowell , David P. Helmbold (1989), "Debugging Concurrent Programs"
    
  • Evolutionary Algorithms
  • Particle Swarms - Kennedy, James and Eberhart, Russell (1995), "Particle Swarm Optimization"
    
  • Genetic Algorithms - Heinz Mühlenbein (1997), "Genetic Algorithms"
    
  • Differential Evolution - Rainer Storn and Kenneth Price (1995), "Differential Evolution - A simple and efficient adaptive scheme for global optimization over continuous spaces" 
    
  • Genetic Programming - Koza, John (1997), "Genetic Programming"
    
  • Machine Learning
  • Reinforcement Learning - Leslie Pack Kaelbling , Michael L. Littman , Andrew W. Moore (1996), "Reinforcement learning: a survey"
  • Supervised vs Unsupervised - Xiaojin Zhu (2006), "Semi-Supervised Learning Literature Survey"
    
  • Pattern Recognition - Anil K. Jain , Robert P. W. Duin , Jianchang Mao (2000), "Statistical pattern recognition: A review"
  • Classification and Segmentation - A. K. Jain , M. N. Murty , P. J. Flynn (1999), "Data Clustering: A Review "
  • Usability studies
    
    
• Begin an initial review of potential sources of information.
  The quality of the references will be used to evaluate your paper, so make sure that the sources you use are high quality, technically rigorous, and reliable.  A web search is a good start, but most of your information should come from respected sources (ACM, IEEE, Papers from these resources provided through CiteSeer / Google Scholar, or other professional societies; books by qualified authors; journals in Business Source Premier, etc.).
• Develop an annotated bibliography of at least 5 sources.
  Cornell has an excellent description of the intent of an annotated bibliography: http://olinuris.library.cornell.edu/ref/research/skill28.htm
• Select a thesis.
• Realize that more depth is expected if you choose to review fewer topics.
  

For the above topics the thesis may be something like:

• X is the preferable evolutionary algorithm for problem type A because of evidence Z...
• Method Y is not a good approach for testing with novice users because ...
• The five most important techniques for developing a software product are…

Comparisons similar to those in news articles such as Tom's Hardware, Wired, Engadget, etc., are unacceptable and will receive a reduced grade.

2. Paper Outline & Peer Review (Due Oct. 3, 7.5 points)

Develop an outline of your planned paper.

You will need a draft outline of your paper that includes introductory information, background on the topics reviewed, comparison information, and conclusion. 

We will do peer review in class, where your outline will be critiqued by at least two of your classmates. You will also be responsible for reviewing the work of at least two of your classmates.

3. Submission

Submit your thesis, annotated bibliography, draft outline, and peer reviews of your outline in a zipped file to the eCompanion Dropbox for this project.

Paper Format:

Required Materials: Title Page, Body, Bibliography

Title Page Contents:

Paper title (Centered on page)
Author (Centered below the title)

Citations: APA style (http://www.library.cornell.edu/resrch/citmanage/apa)
Length: 6-8 pages (not including title and bibliography)
Font: 12 point Times New Roman
Margins: 1 inch margins on all sides
Spacing: Double spaced
Header:

The page title should be in the middle of the header.
The page number should be on the right.
No footer necessary.

Resources:

For an overview of the process of writing a research paper see: http://www.bydewey.com/1steps.html

The Writing Center (http://catpages.nwmissouri.edu/m/WC/selfhelp.html) has self-help resources as you develop your paper.

The KCP&L Research Paper Competition

      This year, Kansas City Power and Light (KCP&L) is sponsoring a research paper competition with prize money.  The amounts have not been finalized for this year, but past year’s prizes totaled $1000 cash for first, second, third, and fourth places. The competition will be based on the research papers that were submitted in 44-411 (Systems Analysis and Design) and 44-561 (Software Engineering I).  A research paper, presentation, and submission to the contest for judging is required in this course.  Failure to be selected for presentation at the contest will not result in a reduced grade.  Participation in the contest is required in one of two ways - a) if selected to present in the contest you must present your paper (preferably using the slides you presented in class) or b) if not selected to present, you must attend at least one presentation and submit a brief written review of that presentation.  Your instructor believes this to be fair.  If you are unable to attend due to work or illness, you must provide written proof such as a doctor's note or work schedule.

Judging Procedures

1. A panel of CS/IS faculty members will rank the submitted papers.  To reduce the chance of bias, the panel will receive papers without the names of the authors.
2. The instructors for 44-411 and 44-561 will not participate in the judging, and the judges will not have access to the grades that were assigned to the papers.
3. On Thursday, Dec. 5, the authors of the top six papers will make a formal presentation of their research to representatives of KCP&L.
4. The final rankings will be determined as follows:
1. The evaluations of the papers by the faculty panel will count 50%
2. The evaluations of the presentations by the KCP&L panel will count 50%
3. Any ties will be broken by the KCP&L panel

Contest Entries
      All students from this class will be entered into the contest and will be required to present their research in class.

1. You are required to submit three printed copies of your revised paper in which your names have been replaced with your 919 numbers.
2. You must prepare, and present in class, a 10-minute presentation based on your research.
   

If you are selected for the contest, you should be able to use the same 10-minute presentation for the contest