Classification of github issues based on natural, aink And fastText libraries

• Autorzy: Mazurek Maciej , Podpora Michal
• Języki publikacji: EN

Abstrakty

EN

The GitHub website issues often have labels wrongly assigned or the labels are missing completely. This article presents a comparison of three libraries in the field of computational linguistics. The goal of this comparison is to select the best solution for the evaluation of algorithms and for minimization of the learning time. The following libraries were used: Natural, Wink and fastText. Each library has been evaluated using a set of 60,000 issues, marked with the labels: bug, enhancement, question.

• Wydawca: Oficyna Wydawnicza Politechniki Opolskiej
• Czasopismo: Zeszyty Naukowe. Elektryka / Politechnika Opolska
• Rocznik: 2019
• Tom: z. 77
• Strony: 73--74
• Opis fizyczny: Bibliogr. 3 poz., rys., tab.

Twórcy

autor

Mazurek Maciej

• Faculty of Electrical Engineering, Automatic Control and Informatics Institute of Computer Science

autor

Podpora Michal

• Faculty of Electrical Engineering, Automatic Control and Informatics Institute of Computer Science

Bibliografia

• [1] KAVIANI P., DHORTE S.: Short Survey on Naive Bayes Algorithm, 2017-11-04, accessed: 2019-05-30, available:https://www.researchgate.net/publication/323946641 _Short_Survey_on_Naive_Bayes_Algorithm/
• [2] Facebook AI Research, Bag of Tricks for Efficient Text Classification, 2016-08-09, accessed: 2019-05-30, available: https://arxiv.org/pdf/1607.01759v3.pdf
• [3] MIKOLOV T., CHEN K., CORRADO G., DEAN J.: Efficient Estimation of Word Representations in Vector Space, 2013-01-16, accessed: 2019-05-30, available: https://arxiv.org/pdf/ 1301.3781.pdf

Uwagi

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).