Multimodal Content Analysis for Effective Advertisements on YouTube

Abstract

The rapid advances in e-commerce and Web 2.0 technologies have greatly increased the impact of commercial advertisements on the general public. As a key enabling technology, a multitude of recommender systems exists which analyzes user features and browsing patterns to recommend appealing advertisements to users. In this work, we seek to study the characteristics or attributes that characterize an effective advertisement and recommend a useful set of features to aid the designing and production processes of commercial advertisements. We analyze the temporal patterns from multimedia content of advertisement videos including auditory, visual and textual components, and study their individual roles and synergies in the success of an advertisement. The objective of this work is then to measure the effectiveness of an advertisement, and to recommend a useful set of features to advertisement designers to make it more successful and approachable to users. Our proposed framework employs the signal processing technique of cross modality feature learning where data streams from different components are employed to train separate neural network models and are then fused together to learn a shared representation. Subsequently, a neural network model trained on this joint feature embedding representation is utilized as a classifier to predict advertisement effectiveness. We validate our approach using subjective ratings from a dedicated user study, the sentiment strength of online viewer comments, and a viewer opinion metric of the ratio of the Likes and Views received by each advertisement from an online platform.

Exploring the opinions of the YouTube visitors towards advertisements and its influence on purchase intention among viewers

“…Also, the number of views could determine how far the ad gained attention on YouTube and the likes it gained validate interest (audio, video, content, and message appeals) based upon the characters persuading the viewers (Santos, 2020). Effectiveness of advertisements on YouTube was explained as a set of multimodal frameworks where framework stands for a joint representation of auditory, visual as well as textual patterns which is developed with the help of multimedia signal processing and natural language processing tools by using the extracts of temporal patterns forming the content of the advertisements (Vedula, Sun, Lee, & Parthasarathy, 2017). Entertainment, informativeness, customization (viewer can customize the information as it suits them) was found to have a positive effect on advertising value (Kelly et al, 2010;Wilson et al, 2012), and irritation had a negative effect where advertising value was defined as a subjective evaluation of value attached to the advertisement (Bang & Lee, 2016;Jung, 2017;Van Doorn & Hoekstra, 2013).…”

Native Advertising in Video Blogs: Language and Genre Specificity

“…behind in the possibilities of communication with real and potential consumers. In the works of Russian and foreign researchers, advertising integrations in video blogs are considered as a highly effective kind of guerrilla marketing [8][9][10][11][12][13], oriented mostly to a young consumer audience [14,15].…”

Enriching Taxonomies With Functional Domain Knowledge

“…Recent literature (e.g. [8,16,39,40,51]) has seen an extensive use of learning and employing high-dimensional word embeddings for diverse applications. We build upon the highly effective Skip-gram variant of word2vec embeddings [23].…”