![[PukiWiki]](image/wadanyan.png "[PukiWiki]")
#author("2019-01-21T19:14:53+09:00","","") #author("2019-02-15T21:24:18+09:00","","") [[論文英語単語帳]] ***リンク [#f20ebad6] #ls2(論文英語単語帳/) *副詞 [#k5e66785] **強調系 [#eb3d5fe4] ***さらに深く [#ge5408f8] -further --To further study the effect of the number of hidden states ***特に [#scc45fc3] -in particular --Previous work studied API for various purposes: (1) to recommend relevant API to developers during development and–or during changes, typically when upgrading the dependencies of one module to newer versions of other modules [21], and (2) to migrate API, in particular in the context of services [22]. -- In particular, versions 5.19 and 5.21 were less failure-prone than their stable successor, while other odd versions were more failure-prone. -Especially --Especially if one OS is less popular than others, it might have seen less testing or some features might not have been ported over, causing test scripts to fail. Such failures would receive less weight in the build results than builds failing consistently across all OSes. ***さらに [#qc91dfc7] -moreover -- Moreover, CPAN has a long history although it provides build data only at the release level. --furthermore --Furthermore, some builds have been performed on older environments but there was not enough build data to study these in details. ***それでも [#ke65c0db] -still --Thus, some sort of mapping must be done to convert the data into a common format that still preserves as much of the meaning of the original data as possible. --While both eects (mediation and moderation) are estimated differently, it is still possible that a variable can have both a mediation and moderation effect [1] (i.e., a mediator M interacts with the relationship between the independent X and the dependent variables Y ). ***はっきり [#rf4b2e0d] -clearly --Love clearly obtains the highest κ agreement, corresponding to a moderate value. ***比較的に [#yf70da46] -relatively --I.e., they are relatively short, written in an informal way (e.g., containing emoticons) and, contrary to regular text parsed in sentiment analysis, they typically contain technical content like stack traces or code snippets interleaved with regular text. ***大幅に [#l46a86a1] -substantially --Compared to functional bugs, performance bugs are substantially more difficult to handle because they often manifest themselves only with large inputs and specific execution environments. --Excluding them from any automatic software engineering techniques could substantially distort the results. ***過度に，あまりにも [#s500ee4c] -overly --developers may use existing performance testing and modeling techniques and do not need to be overly concerned about dealing with configurations. ***とても，猛烈に [#q774ac9b] -tremendously --The main take-home message is that, while release engineering technology has flourished tremendously due to industry, empirical validation of best practices and the impact of the release engineering process on (amongst others) software quality is largely missing and provides major research opportunities. -hugely --Firefox is a hugely popular web browser that has shifted from the traditional development model to a rapid release model. ***理路整然として，明確に [#t4377fd1] -coherently --So instead one must choose a set of discrete categories and devise mapping rules to convert the data consistently and coherently into the new categories. **頻度系 [#sb666db3] ***まれに [#ye958301] -rarely --the javadoc of a class often contains only descriptions of its methods and rarely has code examples on the usage of its objects and methods. ***しばしば [#e31f70c5] -frequently --The application can frequently interact with other system components such as shared libraries, environment variables. **その他 [#d55865bb] ***適切に [#hd854f7f] -properly --Such models are essential to properly manage the major software development outcomes (e.g., cost, schedule, quality) and to build up organizational knowledge. ***わずかに [#y499b060] -slightly -- we constructed a new set of defect categories at a slightly higher level of abstraction than the historical ones. ***実際に [#y4cc159a] -Indeed --Indeed, mobile Apps classification is not a trivial task --On the other hand, if performance bugs are indeed sensitive to configurations, it is also worth exploring the characteristics of such configurations. ***不幸なことに [#n8a84196] -unfortunately -- Unfortunately, API documentation is often insufficient. ***一貫して [#x5f977ad] -consistently --So instead one must choose a set of discrete categories and devise mapping rules to convert the data consistently and coherently into the new categories. --Such failures would receive less weight in the build results than builds failing consistently across all OSes. ***必然的に [#nc45171e] -inevitably --Unfortunately, floating point requires navigating many subtle tradeoffs and inevitably introduces semantics that differ from those of the reals. ***相互に [#f81338f2] -mutually --Some categorizations are orthogonal, i.e., they address different dimensions of interest (e.g., severity vs. technical cause), so combining them results in categories that are not mutually exclusive. ***きちんと，丁寧に [#becbadf7] -scrupulously -- The key to increasing confidence in our guesses was to, first, scrupulously document our categorization decisions. ***直感的に [#b2f2551f] -intuitively --In this area, the need to derive tractable measurements that are reasonable to undertake and intuitively plausible cannot be understated. --Intuitively, if there are many bugs reported in the last x days with a higher severity level than BR, BR is likely not assigned a high priority level since there are many higher severity bug reports in the bug tracking system that need to be resolved too. ***入念に [#e4c54af8] -elaborately --Many labs do not have their defect data elaborately categorized by defect type. ***容易に [#tb19f346] -easily --The complexity of the configuration space and the sophisticated constraints among configuration settings could easily cause performance bugs. ***無期限に [#p6e73ccb] -indefinitely --Since no listening sockets are available, the calling child process waits indefinitely. ***プログラムで [#e1b6c85a] -programmatically --we also utilized the APIs that have been provided to programmatically manipulate internal data structures. ***均一に [#g92fea7a] -homogeneously --Although every new run on a new platform can bring new information, the amount of test results across environments and OSes is not homogeneously distributed, which makes some failures to have more weight than others. ***おそらく [#u541599e] -arguably（十分論証できることだが，おそらく） --One's online social footprint is arguably easier to ascertain and gather than one's technical contributions. -presumably --The communication activities that follow the initiation time would presumably be less cohesive, and more connected to individual preferences and styles of work. ***故意に [#neea084d] -deliberately --In our case, the normal severity is deliberately not taken into account. ***に渡って [#i908b573] -across --These numbers are more or less stable across the five cases, with some fluctuations. --Since the comments are randomly sampled across all projects’ comments, the more issue comments a project has, the higher the probability that we analyzed some of its comments in the sample, i.e., large projects are represented more in the data set. ***逆に [#wdb383eb] -Conversely --Conversely, the smaller the project, the lower the chance that the analyzed sample has comments of such project, if any. ***この理由で [#zf8af20c] -Hence --Hence, this RQ aims to understand how often environments break the build and to what degree a failing build can recover again or is doomed to keep on failing. --Hence, development releases are used less for builds and have less reliable build results. ***一般的に [#paee941d] -Typically --Typically, these models employ some functions of software design and code metrics to predict faults, or fault-proneness [2, 26]. ***そのかわりに [#g87c9da6] -Alternatively --In contrast to prior studies on fault-proneness, the mediation effect is not evident across all systems examined. Alternatively, we found that 2) there is more evidence that size has a significant moderation effect, especially for CBO and Fan-out metrics, as it impacts the strength/magnitude of the relationship between OO metrics and faults. ***その後 [#w78793a2] -Subsequently --Subsequently, we conduct moderation analysis by building a regression model to model-fit the outcome (number of faults) using the independent variable (an OO metric), the moderator and an interaction term which tests whether the relationship between the OO metric and outcome changes based on the moderator. ***公開されている [#id8ff743] -publicly --To examine the possible size effect we selected two publicly avail- able open source datasets from D’Ambros et al. [5] and Jureczko and Madeyski [16]. ***厳密に [#ke070dc0] -rigorously --But how is test code reviewed? Is it reviewed as rigorously as production code, or is it reviewed at all? Are there specific issues that reviewers look for in test files?
