Judging YouTube by its Covers
نویسنده
چکیده
In this new age of digital multimedia content and internet audiences, the entertainment industry and the path to stardom is slowly evolving. Many aspiring musicians can now achieve fame by performing the works of other famous artists, and sharing their performances on sites such as Youtube. Aspiring artists and other content generators on Youtube are always advised by marketing experts to include keywords in their video titles, descriptions and tags. In this paper, we show that this advice is sound. Using only the title of the video and the Bernoulli Naive Bayes model, we predict whether a video receives many or few views with 65.83% accuracy. We also show that the Bernoulli Naive Bayes model performs better than SVMs and k-NN on this data set of music cover videos.
منابع مشابه
American Virtual Colonialism and the Islamophobia Politics: Muslim/Iranian Women’s “Hijab” at “YouTube”
Virtual Colonialism (Ameli, 2011) is reflecting a new trend of colonialismin the virtual space parallel to the physical space. According to Ameli(2007), withthe “Dual Spacization” of life,the imperial powers like the United States of Americatake the advantage of the synchronization and translocality of the virtual space toexpand their realm of imperial power through the “attraction” and “persua...
متن کاملYouDACC: the Youtube Dialectal Arabic Comment Corpus
In the Arab world, while Modern Standard Arabic is commonly used in formal written context, on sites like Youtube, people are increasingly using Dialectal Arabic, the language for everyday use to comment on a video and interact with the community. These user-contributed comments along with the video and user attributes, offer a rich source of multi-dialectal Arabic sentences and expressions fro...
متن کاملUnder the Covers of IEEE Software
ECOSYSTEMS THAT THRIVE are those whose members contribute more than they take away. Judging by the growth of its offerings and volunteers, IEEE Software is clearly such a case: over the past year, more than 1,500 people have contributed to it as authors, reviewers, editors, podcast hosts and guests, advisors, curators, and more. What makes Software tick, and how can you participate in its commu...
متن کاملUnderstanding the Characteristics of Category-Specific YouTube Videos
As the world largest video content sharing website, YouTube constantly attracts more and more attention from networking research community. Different aspects of YouTube, such as video characteristics, user behaviors, and its back-end infrastructure have been well studied. Most of these studies consider YouTube as a whole without taking the YouTube category into the consideration. However, it is...
متن کاملJudging a Book By its Cover
Book covers communicate information to potential readers, but can that same information be learned by computers? We propose using a deep Convolutional Neural Network (CNN) to predict the genre of a book based on the visual clues provided by its cover. The purpose of this research is to investigate whether relationships between books and their covers can be learned. However, determining the genr...
متن کامل