
A/B-теÑтирование — метод иÑÑледованиÑ, Ñуть которого заключаетÑÑ Ð² том, чтобы выÑÑнить, какие из изменений Ñлементов ПО улучшают целевой показатель теÑта. Метод помогает принимать Ñ€ÐµÑˆÐµÐ½Ð¸Ñ Ð½Ð° оÑнове реальных данных, отражающих Ð¿Ñ€ÐµÐ´Ð¿Ð¾Ñ‡Ñ‚ÐµÐ½Ð¸Ñ Ð¿Ð¾Ð»ÑŒÐ·Ð¾Ð²Ð°Ñ‚ÐµÐ»ÐµÐ¹ ПО. Работа раÑкрывает методологичеÑкие аÑпекты Ð¿Ñ€Ð¾Ð²ÐµÐ´ÐµÐ½Ð¸Ñ A/B теÑтов, вÑех ключевых Ñтапов – от Ð¿Ð»Ð°Ð½Ð¸Ñ€Ð¾Ð²Ð°Ð½Ð¸Ñ Ð¸ Ð¿Ñ€Ð¾Ð²ÐµÐ´ÐµÐ½Ð¸Ñ ÑкÑперимента до анализа результатов теÑта. Цель работы – разработка методологии полного цикла Ð¿Ñ€Ð¾Ð²ÐµÐ´ÐµÐ½Ð¸Ñ Ð/B ÑкÑпериментов в мобильном приложении. Подцелью работы ÑвлÑетÑÑ Ñ€Ð°ÑÑмотрение Ð/B теÑтов Ñ Ð¸ÑторичеÑкой точки Ð·Ñ€ÐµÐ½Ð¸Ñ Ñ€Ð°Ð·Ð²Ð¸Ñ‚Ð¸Ñ Ñ€Ð°Ð½Ð´Ð¾Ð¼Ð¸Ð·Ð¸Ñ€Ð¾Ð²Ð°Ð½Ð½Ñ‹Ñ… ÑкÑпериментов и ÑтатиÑтичеÑких методов обработки данных до Ñовременных проблем и подходов к Ð/B ÑкÑпериментам. Ð’ работе также были раÑÑмотрены Ñовременные платформы Ð´Ð»Ñ Ð¿Ñ€Ð¾Ð²ÐµÐ´ÐµÐ½Ð¸Ñ ÑкÑпериментов в мобильных приложениÑÑ…. Ð’ практичеÑкой чаÑти был Ñпланирован и проанализирован A/B теÑÑ‚ в мобильном приложении.
A/B testing is a research method, the essence of which is to find out which of the changes in the software elements improve the test target. The method helps to make decisions based on real data reflecting the preferences of software users. The work reveals the methodological aspects of conducting A/B tests, all key stages – from planning and conducting an experiment to analyzing test results. The purpose of the work is to develop a methodology for a full cycle of A/B experiments. The sub-goal of the work is to consider A/B tests from the historical point of view of the development of randomized experiments and statistical methods of data processing to modern problems and approaches to A/B experiments with a narrow specialization in the field of conducting randomized experiments in mobile applications. The article also considered modern platforms for conducting experiments in mobile applications. In the practical part, an A/B test was planned and analyzed in a mobile application.
дизайн ÑкÑпеÑименÑа, пÑовеÑка гипоÑез, multiple testing, а/б ÑеÑÑиÑование, мобилÑнÑе пÑиложениÑ, множеÑÑвенное ÑеÑÑиÑование, пÑинÑÑие ÑеÑений, randomized experiments, mobile applications, decision making, statistical criteria, ÑандомизиÑованнÑе ÑкÑпеÑименÑÑ, a/b testing, ÑÑаÑиÑÑиÑеÑкие кÑиÑеÑии, hypothesis testing, experiment design
дизайн ÑкÑпеÑименÑа, пÑовеÑка гипоÑез, multiple testing, а/б ÑеÑÑиÑование, мобилÑнÑе пÑиложениÑ, множеÑÑвенное ÑеÑÑиÑование, пÑинÑÑие ÑеÑений, randomized experiments, mobile applications, decision making, statistical criteria, ÑандомизиÑованнÑе ÑкÑпеÑименÑÑ, a/b testing, ÑÑаÑиÑÑиÑеÑкие кÑиÑеÑии, hypothesis testing, experiment design
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
