论文信息 - Mining Helium programs with Neon

Mining Helium programs with Neon

Over the years we have collected a large collection of Haskell programs written by students in a first-year functional programming course using the Helium compiler. The mining of such a collection is not trivial, especially since the programming was done in vivo, and hence largely outside our control. We have developed a sizable library in Haskell, called Neon, for computing characteristics of this collection of programs and presenting the results visually. These computations range from simple kinds of analyses like computing the average length of a program per student to determining how long it takes for a programmer to resolve a type error.

J. Hage | P. Keeken

[1] P. G. Moulton,et al. DITRAN—a compiler emphasizing diagnostics , 1967, CACM.

[2] J. J. Horning,et al. The impact of language design on the production of reliable software , 1975 .

[3] Gordon B. Davis,et al. A study of errors, error-proneness, and error diagnosis in Cobol , 1976, CACM.

[4] Marvin V. Zelkowitz. Automatic program analysis and evaluation , 1976, ICSE '76.

[5] Stef Joosten,et al. Teaching functional programming to first-year students , 1993, Journal of Functional Programming.

[6] Daan Leijen,et al. Helium, for learning Haskell , 2003, Haskell '03.

[7] Jurriaan Hage,et al. Scripting the type inference process , 2003, ACM SIGPLAN Notices.

[8] Matthew C. Jadud. A First Look at Novice Compilation Behaviour Using BlueJ , 2005, Comput. Sci. Educ..

[9] Simon J. Thompson,et al. Software metrics: measuring Haskell , 2005, Trends in Functional Programming.

[10] Master Thesis,et al. Analyzing Helium Programs Obtained Through Logging - The process of mining novice Haskell programs - , 2006 .

[11] Jurriaan Hage,et al. Heuristics for type error discovery and recovery (revised) , 2006 .

[12] Jurriaan Hage,et al. Mining for Helium , 2006 .

[13] Jurriaan Hage,et al. Heuristics for Type Error Discovery and Recovery , 2006, IFL.