Video content recommendation: An overview and discussion on technologies and business models

Abstract

This paper presents several aspects, from technologies to business models, of content recommendation for video related solutions that range from IP television (IPTV), including linear programming television (LPTV) and videoon-demand (VoD), to online video. Video content recommendation is becoming increasingly important because of the continuously increasing amount of video content available to end users. After considering some end user requirements, an analysis is provided of the most important content recommendation technologies as described in literature and implemented by many start-ups. The paper also deals with evaluation criteria for a content recommender system related to user expectations, support of different scenarios (e.g., new content, new users), and marketing and business requirements. We describe the overall architecture into which the content recommendation functionality fits, as well as its interfaces with other network components, external databases, social networks, and applications.Finally, we discuss a dedicated network provider play, the business opportunities, and several business models for content recommendation. © 2011 Alcatel-Lucent. very exciting, the overwhelming choice in content available on broadcast TV, VoD, and the Internet is also leading to much irritation and frustration of the end user, since it is becoming increasingly hard to pick out that one movie, program, or video clip best suited to his mood, environment, and interest at that moment. It demands quite some patience, perseverance, and time from the end user to sift the programming guide, click through all channels, or browse the complete VoD catalog in order to find something of interest to watch. Because of this, people tend to stick IntroductionThe expansion of the Internet and the advent of digital broadcasting have led to an ever-increasing range of content at the fingertips of the end user. Video-on-demand (VoD) libraries are approaching 20,000 titles per aggregator with substantial further growth expected; the number of online aggregators is increasing continuously; millions of Internet video clips are accessible on television (TV); and the number of linear programming television (LPTV) channels is increasing constantly, along with content on catchup TV (recordings of LPTV). Although this sounds to what they are used to watching, and the promise of having everything you like at your fingertips is still only a promise.This dilemma not only short-circuits the end user experience, but sets the stage for dissatisfaction and the risk that an end user will turn his back on the service as well as the content provider. An NDS Group survey of more than 1,000 cable customers in the United States (U.S.) found that LPTV program recommendations are one of the top applications they desire. Keeping in mind that millions of people in the U.S. are spending 150 hours per month in front of their TV sets [19], improving the end user experience with content discovery can lead to a win-win-win situation for the end user, service prov...

Optimizing for Video Storage Networking With Recommender Systems

“…We also introduce the parameter b, which indicates the percentage increase of the overall number of requests. A typical value of b, which we also call the boost factor, is around 15 percent as in [3,9]. b depends on g.…”

“…An end-to-end service typically provides both discovery and delivery functionality which might be seen as belonging to horizontal networking layers as discussed in [9]. A vertical split in (autonomous) responsibility domains facilitates partitioning proper functionality with the objective of increasing the end user's QoE, expressed in terms of 1) the ease in finding content of interest, 2) the ability to watch selected content instantly, and 3) the availability of the content in high definition (HD) video quality.…”

“…One of the reasons is that there is a need for sufficient requests and ratings both from end users and for items over a sufficiently long period. In other words, the rating matrix over the time period of interest must not be too sparse (see [11] for collaborative filtering and [1,9] for content based recommenders). Additionally a recommender typically requires highly complex and extensive computations and economizing on this will lead to approximations.…”

“…With respect to discovery, customers need help in finding content that interests them, especially as the amount of available content increases. Therefore, recommender systems are often integrated with video offerings to help the end user [1,3,9,13]. The use of recommenders for the purpose of caching has also been discussed in [10].…”

Research on the fusion mechanism of cooperative embedded filtering and crowd content recommendation

“…The content recommendation for video-related solutions was proposed in article [14], which ranged from IP television. A M-Learning Content Recommendation Service was provided by exploiting the mobile social interactions in article [15].…”