『독립신문』 논설의 형태 주석 말뭉치를 활용한 논설 저자 판별 연구

This paper tries to build Doknipshinmun corpus including active morph tags and analyze an authorship attribution of the Editorials in Doknipshinmun using this corpus. Based on the computerized corpus data set consisting of 212 anonymous Editorials, the approach utilizes relative frequencies of lexical unit, such as inflected word-endings in Korean in an attempt to discriminate authorship between Dr. Phillip So and Ju Shi-kyung. The basic assumption of this analysis is that the author of an editorial can be selected from a set of possible authors by comparing the values of lexical characteristic measurements in that text to their corresponding values in each possible author’s known writing sample. Using 4 known editorials for Dr. Phillip So and Ju Shi-kyung respectively as a calibrate sample, the results indicate that quantitative vector space model technique in authorship attribution and computational stylistics can be successfully applied to identify the authors of Doknipshinmun Editorials. Some limitations of quantitative authorship study are discussed.