![[PukiWiki]](image/wadanyan.png "[PukiWiki]")
#author("2018-07-30T02:58:19+09:00","","") #author("2018-07-30T03:00:24+09:00","","") [[論文英語単語帳/表現集]] ***リンク [#f20ebad6] #ls2(論文英語単語帳/) ** 表現集[#s089e92b] #ls2(論文英語単語帳/表現集/) *サンプル [#xe17330c] **リサーチクエスチョン内（特に書き始め） [#e98eb1ac] *リサーチクエスチョン内 [#b72bac40] **起 [#e98eb1ac] -want --The interpretation of a phrase usually depends on the previous discussion (i.e., context) of the conversation (Tepperman et al 2006). For example, the sentence “yeah, right” can have a different meaning (both sarcastically and otherwise) when following a sentence like “with java 8 we fix all problems” versus “breaking backward compatibility is risky”. For this reason, rating a comment without its context can be compared to eavesdropping on a group conversation and only catching the last phrase of the conversation. However, due to the technical and unstructured nature of software development artifacts, the impact of context might be different in technical prose than in normal language. Here, we want to analyze the impact of context on agreement between raters. -challenge --Therefore, identifying emotional content in technical prose within software development artifacts is a challenge, let alone agreeing on this content between different human raters. -aim --This initial research question aims at understanding how often builds fail and whether the ratio of failing builds is a constant value or fluctuates across time. So we investigate build inflation in terms of number of builds. -seem --Build inflation seems to correspond to differences/similarities in OSes/environments, which can identify clusters of Perl versions/OSes with different failure ratio and value. T --Build inflation seems to correspond to differences/similarities in OSes/environments, which can identify clusters of Perl versions/OSes with different failure ratio and value. **承 [#r973c5db] -to explain --To explain the decrease of the build failure ratio observed in RQ1, our hypothesis is that a given new distversion is built multiple times in such a way that most of these builds succeed, while only few fail.
