Video Content Popularity Research Articles

Intelligent Caching for Mobile Video Streaming in Vehicular Networks with Deep Reinforcement Learning

Caching-enabled multi-access edge computing (MEC) has attracted wide attention to support future intelligent vehicular networks, especially for delivering high-definition videos in the internet of vehicles with limited backhaul capacity. However, factors such as the constrained storage capacity of MEC servers and the mobility of vehicles pose challenges to caching reliability, particularly for supporting multiple bitrate video streaming caching while achieving considerable quality of experience (QoE). Motivated by the above challenges, in this paper, we propose an intelligent caching strategy that takes into account vehicle mobility, time-varying content popularity, and backhaul capability to improve the QoE of vehicle users effectively. First, based on the mobile video mean opinion score (MV-MOS), we designed an average download percentage (ADP) weighted QoE evaluation model. Then, the video content caching problem is formulated as a Markov decision process (MDP) to maximize the ADP weighted MV-MOS. Owing to the prior knowledge of video content popularity and channel state information that may not be available at the road side unit in practical scenarios, we propose a deep reinforcement learning (DRL)-based caching strategy to solve the problem while achieving a maximum ADP weighted MV-MOS. To accelerate its convergence speed, we further integrate the prioritized experience replay, dueling, and double deep Q-network technologies, which improve the performance of DRL algorithm. Numerical results demonstrate that the proposed DRL-based caching strategy significantly improves QoE, and achieves better video delivery reliability compared to existing non-learning approaches.

동영상 플랫폼상의 식품 광고 타겟팅의 효과: PC와 모바일 비교를 중심으로*

1. Introduction With the increasing popularity of video contents, companies are keen on utilizing video content as an advertising space. Targeted advertising is directed towards consumers with certain characteristics (Plummer et al. 2007), and food advertisements are usually exposed to consumers of a certain age at specific times along with popular TV programs favored by the target segment (Soni and Vohra 2014). As such, targeted advertising has received a lot of attention as an effective way for companies to reach potential customers who are more likely to purchase; however, research on targeted advertising still lacks. Therefore, this study focuses on various targeting strategies widely used in video advertisements and examines their effects on advertising effectiveness in the food industry. This study further investigates that these targeting effects vary between devices. Specifically, this study intends to answer the following research questions. First, how do the targeting strategies affect the effectiveness of food advertisement on the video platform? Second, how does the device type moderate the aforementioned effects? 2. Literature Review and Hypotheses Development Companies today have a lot of information on consumer preferences and media habits (Iyer, David and Miguel 2005), so targeted advertising can be executed to increase the interest of specific consumers, thereby increasing advertising effectiveness. Currently, there are three general targeting strategies on video platform: 1) contextual targeting, 2) popular content targeting, and 3) demographic and time targeting which can be classified as audience targeting. According to various targeting strategies, consumers view advertisements with different degrees of relevance to themselves when consuming media. As for devices, PC and mobile have some differences in internet usage. PC is effective for multitasking and thus PC users are exploration-oriented, so new information and interest provided by advertisements are important factors in determining ad completion. Exploration-oriented users are highly open to information that they did not know, and thus show relatively high interest and openness to advertisements that are not relevant to them. However, mobile is less convenient for multitasking than PC and thus mobile users are goal-orientated, so not only new information and interest provided by advertisements but also perceived intrusiveness caused by advertisements are important factors in determining ad completion. Goal-oriented users focus more on their tasks, and thus show relatively low interest and openness to advertisements that are not relevant to them. Based on the characteristics of targeting strategies and devices, we hypothesize the interplay of targeting strategies and devices on the effectiveness of food advertisement as follows: [H1] When watching video content on PC, popular content targeting in comparison with contextual targeting has a positive effect on ad completion. [H2] When watching video content on PC, audience targeting in comparison with contextual targeting has a positive effect on ad completion. [H3] When watching video content on mobile, the PC-based positive effect of popular content targeting compared to contextual targeting decreases. [H4] When watching video content on mobile, the PC-based positive effect of audience targeting compared to contextual targeting decreases. 3. Data and Analysis To verify the research hypothesis discussed above, the authors obtained users

Застосування інструментів цифрового маркетингу для підвищення ефективності підприємницької діяльності

Abstract. Introduction. The advantages of digital marketing include better audience coverage, the ability to collect clear and detailed data, flexible approach, interactivity, lack of territorial restrictions in the implementation of marketing ideas, easy access to resources, real-time marketing management. Purpose. The purpose of the study is to substantiate and develop theoretical and practical principles of using digital marketing tools to improve business efficiency. Results. Based on the analysis of the definitions of "digital marketing", our own definition is proposed: it is the marketing of goods and services through digital channels and tools in virtual and real environments for effective interaction with existing and potential consumers. The main trends in digital marketing include the growing popularity of video content, social networks and messengers such as promotion channels, multi-channel marketing, working with "big data", mobile content and more. Based on the review of trends in educational content and requests of the generation of "millennials" for information, it was found that the medical center "Gold Clinic" should develop its digital information platforms in the direction of filling with relevant content, providing different formats of content, providing it in real time time. Conclusions. Gold Clinic lags behind in terms of number of views, number of pages viewed and time spent on the site, as well as in terms of visibility on the Internet. Therefore, now content marketing and search engine optimization should become strategic priorities to improve the efficiency of digital marketing of the enterprise. Thus, in order to personalize the content for clients of the Gold Clinic, it is advisable to use modern technologies such as artificial intelligence, machine learning, decisionmaking based on the analysis of "big data", augmented reality and more.

VIDEO AND AUDIO CONTENT ARE EFFECTIVE COMPONENTS OF SOCIAL NETWORK DEVELOPMENT

The questions of the necessity to develop a social network are considered. The developed social network is offered for publishing new posts, adding and removing other members from the friends list, personal correspondence with any member. Particular attention is paid to the issues of using video content and the reasons for the popularity of video content are determined. A simple and user-friendly design is proposed to accomplish the task. The functions of the user in the developed system are considered and the use cases diagram is constructed. The social network database is described. The choice of the PHP programming language for the task implementation is analyzed and substantiated. Features of use and advantages of Open Server and Bootstrap framework are considered. The use in the development of the social network Ajax - the technology of developing web-applications has been substantiated. The work of the developed social network on the part of the user and the administrator are described. Particular attention is paid to the authorization of the user on the network, for the safety of pages. The test results are presented to check the system's correctness and the compliance of its implementation with the requirements. The project has been uploaded to shared hosting, which makes the social network ready for use on the Internet

The Fission of the Forbidden: The Popularity of Video Content in an Online Right-Wing Extremist Environment

Abstract Video platforms such as YouTube provide an environment where the blurred duality between content dissemination and creation facilitates the generation of social networks. Research into online violent extremist environments has often noted the prominence of video-sharing platforms as a means of distributing propaganda and cultivating social networks for purposes of recruitment. This paper draws from the study of emotion to examine three videos and associated comments that had high engagement, understood as the frequency of interactions, likes/upvotes and reposts in a given social network, in a right-wing extremist online milieu. This analysis highlights the important role emotions play in generating social connectedness and ultimately engagement and recruitment into online right-wing extremist milieus. Understanding the significance of emotions in online violent extremist video content can help to identify opportunities for moderation and/or the construction of alternative narratives.

Hybrid machine learning approach for popularity prediction of newly released contents of online video streaming services

In the industry of video content providers such as VOD and IPTV, predicting the popularity of video contents in advance is critical, not only for marketing but also for network usage. By successfully predicting user preferences, contents can be optimally deployed among servers which ultimately leads to network cost reduction. Many previous studies have predicted view-counts for this purpose. However, they normally make predictions based on historical view-count data from users, given the assumption that contents are already published to users. This can be a downside for newly released contents, which inherently does not have historical data. To address the problem, this research proposes a hybrid machine learning approach for the popularity prediction of unpublished video contents.In this paper, we propose a framework which effectively predicts the popularity of video contents, via a combination of various methods. First, we divide the entire dataset into two types, according to the characteristics of the contents. Next, the popularity prediction is performed by either using XGBoost or neural net with category embedding, which helps resolving the sparsity of categorical variables and requiring the system to learn efficiently for the specified deep neural net model. In addition, we use the FTRL model to alleviate the volatility of view-counts. Experiments are carried out with a dataset from one of the top streaming service companies, and results display overall better performance compared to various standalone methods.

Cooperative and efficient content caching and distribution mechanism in 5G network

This paper considers encoding caching and collaborative distribution issues jointly, and proposes a collaborative and efficient content caching and distribution mechanism; and based on the proposed scheme, an energy consumption model is established from the perspective of content caching and content distribution to optimize Energy efficiency of content caching and content distribution. In addition, this paper proposes a cache placement algorithm based on heuristic greedy algorithm to solve this energy efficiency optimization problem by optimizing cache placement. Finally, the simulation results show the performance of the scheme. This paper proposes a content clustering distribution scheme for the problems of random user distribution and limited cache. This solution uses the user’s media cloud and communication technology to complete the distribution of video content. In addition, the use of an efficient adaptive media cloud clustering mechanism and a minimized hops algorithm has enabled the reasonable deployment of 5G network caches based on the popularity of video content and user distribution. The simulation results show that the system content request response time is shortened by 50%, which effectively reduce the impact of content popularity changes on system performance. This paper analyses the important performance indicators such as cache hit rate, optimal communication radius, and maximum link access in 5G networks for multi-layer popular multimedia content distribution in response to the heterogeneity and transmission interference problems for 5G networks in 5G communications. Under the premise of noise ratio and signal-to-interference ratio threshold, related service link selection and scheduling algorithms is further proposed obtaining simulation results that approximate theoretical analysis, which effectively reduces the number of interference by 40% and improves the system throughput by 10%.

Factors Affecting the Popularity of Video Content on Live-Streaming Services: Focusing on V Live, the South Korean Live-Streaming Service

Naver V Live, a South Korean live-streaming service, showcases video contents specific to the entertainment industry, such as K-pop and music. On V Live, K-pop stars and their fans can interact directly in a natural way, and V Live provides high-quality video content with novel topics. This study has identified key characteristics of video content that affect its popularity. A total of 620 video contents of five leading Star channels were classified on the basis of production company, type of video content, and whether it was live-streamed or not. The popularity of video content was measured by the number of comments, hearts, and views. To control potential bias, additional variables were set as control variables—such as the number of channel subscribers, mini-album sales, if the video content was previewed, and cumulative number of days since the video content was uploaded. For analysis, a hierarchical linear regression was conducted. The findings suggest future directions in video content planning.

Real-Time Algorithms for the Detection of Changes in the Variance of Video Content Popularity

As video content is responsible for more than 70% of the global IP traffic, related resource allocation approaches, e.g., using content caching, become increasingly important. In this context, to avoid under-provisioning, it is important to rapidly detect and respond to changes in content popularity dynamics, including volatility, i.e., changes in the second order moment of the underlying process. In this paper, we focus on the early identification of changes in the variance of video content popularity, which we address as a statistical change point (CP) detection problem. Unlike changes in the mean that can be well captured by non-parametric statistical approaches, to address this more demanding problem, we construct a hypothesis test that uses in the test statistic both parametric and non-parametric approaches. In the context of parametric models, we consider linear, in the form of autoregressive moving average (ARMA), and, nonlinear, in the form of generalized autoregressive conditional heteroskedasticity (GARCH) processes. We propose an integrated algorithm that combines off-line and on-line CP schemes, with the off-line scheme used as a training (learning) phase. The algorithm is first assessed over synthetic data; our analysis demonstrates that non parametric and GARCH model based approaches can better generalize and are better suited for content views time series with unknown statistics. Finally, the non-parametric and the GARCH based variations of our proposed integrated algorithm are applied on real YouTube video content views time series, to illustrate the performance of the proposed approach of volatility change detection.

Multi-node D2D communications for wireless video delivery over cellular networks

Video service is one of the popular services among mobile users, which plays an important role in network traffic growth and owns the largest data traffic compared to other types of communications. Over the last few years, users’ considerable enthusiasm to watch live video streams has captivated the attention of researchers to the networks equipped with content saving devices. In these devices the video content is saved beforehand within the edge devices (small stations and smart devices) for fast delivery. In this study, we propose a new method for sending video content, based on Device-to-Device (D2D) communications and providing multi-relational and multi-user opportunities. Based on the random nature of the popularity of video content and the random requests from users on the network, the authors investigate the probability of sending content to multiple users using the random distribution of popularity of users simultaneously. In this method, several users’ requests are responded by a single transmission. Extensive simulations result of the proposed method reveals the significant effect of the random distribution of video content popularity and shows that using multi-link D2D connections can cause considerable improvements in the efficiency of the network in terms of the network throughput.

Temporal Video Analyzer (TVAN): Efficient Temporal Video Analysis for Robust Video Description and Search

With the increasing popularity of video content, automatic video understanding is becoming more and more important for streamlining video content consumption and reuse. In this work, we present TVAN—temporal video analyzer—a system for temporal video analysis aimed at enabling efficient and robust video description and search. Its main components include: temporal video segmentation, compact scene representation for efficient visual recognition, and concise scene description generation. We provide a technical overview of the system, as well as demonstrate its usefulness for the task of video search and navigation.

Real-Time Video Content Popularity Detection Based on Mean Change Point Analysis

Video content is responsible for more than 70% of the global IP traffic. Consequently, it is important for content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. In this paper, we propose the employment of on-line change point (CP) analysis to implement real-time, autonomous and low-complexity video content popularity detection. Our proposal, denoted as real-time change point detector (RCPD), estimates the existence, the number and the direction of changes on the average number of video visits by combining: (i) off-line and on-line CP detection algorithms; (ii) an improved time-series segmentation heuristic for the reliable detection of multiple CPs; and (iii) two algorithms for the identification of the direction of changes. The proposed detector is validated against synthetic data, as well as a large database of real YouTube video visits. It is demonstrated that the RCPD can accurately identify changes in the average content popularity and the direction of change. In particular, the success rate of the RCPD over synthetic data is shown to exceed 94% for medium and large changes in content popularity. Additionally, the dynamic time warping distance, between the actual and the estimated changes, has been found to range between 20 samples on average, over synthetic data, to 52 samples, in real data. The rapid responsiveness of the RCPD is instrumental in the deployment of real-time, lightweight load balancing solutions, as shown in a real example.

Inter-Cluster Cooperation for Wireless D2D Caching Networks

Proactive wireless caching and device to device (D2D) communication have emerged as promising techniques for enhancing users’ quality of service and network performance. In this paper, we propose a new architecture for D2D caching with inter-cluster cooperation. We study a cellular network in which users cache popular files and share them with other users either in their proximity via D2D communication or with remote users using cellular transmission. We characterize the network average delay per request from a queuing perspective. Specifically, we formulate the delay minimization problem and show that it is NP-hard. Furthermore, we prove that the delay minimization problem is equivalent to the minimization of a non-increasing monotone supermodular function subject to a uniform partition matroid constraint. A computationally efficient greedy algorithm is proposed which is proven to be locally optimal within a factor $(1 - e^{-1})\approx 0.63$ of the optimum. We analyze the average per request throughput for different caching schemes and conduct the scaling analysis for the average sum throughput. We show how throughput scaling depends on video content popularity when the number of files grows asymptotically large. Simulation results show a delay reduction of 45% to 80% compared to a D2D caching system without inter-cluster cooperation.

Scaling Behavior for Device-to-Device Communications With Distributed Caching

We analyze a novel architecture for caching popular video content to enable wireless device-to-device (D2D) collaboration. We focus on the asymptotic scaling characteristics and show how they depend on video content popularity statistics. We identify a fundamental conflict between collaboration distance and interference and show how to optimize the transmission power to maximize frequency reuse. Our main result is a closed form expression of the optimal collaboration distance as a function of the model parameters. Under the common assumption of a Zipf distribution for content reuse, we show that if the Zipf exponent is greater than 1, it is possible to have a number of D2D interference-free collaboration pairs that scales linearly in the number of nodes. If the Zipf exponent is smaller than 1, we identify the best possible scaling in the number of D2D collaborating links. Surprisingly, a very simple distributed caching policy achieves the optimal scaling behavior.