This proposal aims to address the problem of source code plagiarism detection in university laboratories. Existing Java source code comparators have a tradeoff between reliability and speed of similarity detection. There is also a need to approximate the grade a student would receive based on the similarity percentage.
The team proposes to enhance Flaggie, an open source software version of Plag, which is the most efficient Java source code comparator according to related literature. There are two possible approaches - combining the algorithm of a fast but less reliable comparator with Flaggie's algorithm, or formulating a new algorithm using tokenizing and other methods to balance reliability and speed. For grade approximation, the team will use the algorithm from a previous year's thesis
Python Machine Learning: Using Scikit Learn, TensorFlow, PyTorch, and Keras, an Introductory Journey into Machine Learning, Deep Learning, Data Analysis, Algorithms, and Data Science
This proposal aims to address the problem of source code plagiarism detection in university laboratories. Existing Java source code comparators have a tradeoff between reliability and speed of similarity detection. There is also a need to approximate the grade a student would receive based on the similarity percentage.
The team proposes to enhance Flaggie, an open source software version of Plag, which is the most efficient Java source code comparator according to related literature. There are two possible approaches - combining the algorithm of a fast but less reliable comparator with Flaggie's algorithm, or formulating a new algorithm using tokenizing and other methods to balance reliability and speed. For grade approximation, the team will use the algorithm from a previous year's thesis
This proposal aims to address the problem of source code plagiarism detection in university laboratories. Existing Java source code comparators have a tradeoff between reliability and speed of similarity detection. There is also a need to approximate the grade a student would receive based on the similarity percentage.
The team proposes to enhance Flaggie, an open source software version of Plag, which is the most efficient Java source code comparator according to related literature. There are two possible approaches - combining the algorithm of a fast but less reliable comparator with Flaggie's algorithm, or formulating a new algorithm using tokenizing and other methods to balance reliability and speed. For grade approximation, the team will use the algorithm from a previous year's thesis
This proposal aims to address the problem of source code plagiarism detection in university laboratories. Existing Java source code comparators have a tradeoff between reliability and speed of similarity detection. There is also a need to approximate the grade a student would receive based on the similarity percentage.
The team proposes to enhance Flaggie, an open source software version of Plag, which is the most efficient Java source code comparator according to related literature. There are two possible approaches - combining the algorithm of a fast but less reliable comparator with Flaggie's algorithm, or formulating a new algorithm using tokenizing and other methods to balance reliability and speed. For grade approximation, the team will use the algorithm from a previous year's thesis
1 st Semester A/Y 2014-2015 SAD System PROPOSAL Proposal Title # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Propo"e"ts#Se$tio"% !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Re&ie'e( )y#(ate% *omme"ts#Remar+s% _________________________ _____________________________ S,ort Des$riptio" o- t,e St.(y% Sorce co!e plagiarism in la"oratories is one of the pro"lems of the professors# There are alrea!y e$isting %a&a sorce co!e comparators' ho(e&er) there is a tra!e off "et(een the relia"ility an! the spee! of similarity !etection# There is also a nee! for appro$imation on (hat gra!e the st!ent (ill recei&e (hich is !ri&en from the similarity percentage# <YOU CAN WRITE AS WELL THE PROBLEMS THAT YOU AIM TO SOLVE> The team !eci!e! to enhance *laggie) an open sorce soft(are &ersion of %plag (hich is the most efficient %a&a sorce co!e comparator accor!ing to the relate! literatre that the team ha! researche! on# There (ill "e t(o possi"le approach to the pro"lem# +irst is com"ining the algorithm of a fast "t less relia"le ,a&a sorce co!e comparator an! com"ining *laggie-s algorithm# Secon! is formlating a ne( algorithm (ith the se of to.eni/ing an! other metho!s to "alance or optimi/e the relia"ility an! spee! of *laggie# 0ith regar!s to the gra!e appro$imation for the st!ent) the team (ill se the algorithm of the 122 Sorce 1o!e 1omparator *lg-in for 3clipse 4a thesis of the pre&ios year5 as a "asis# Area o- Spe$iali/atio"%___Systems De&elopment Tools#Pro0rammi"0 La"0.a0e to )e .se(6___%a&a__________________________ _________________________________
Python Machine Learning: Using Scikit Learn, TensorFlow, PyTorch, and Keras, an Introductory Journey into Machine Learning, Deep Learning, Data Analysis, Algorithms, and Data Science