22-11-2012, 06:00 PM
Challenges in a Web-enhanced Personalised IPTV Service
Challenges in a Web-enhanced.pdf (Size: 864.6 KB / Downloads: 33)
ABSTRACT
Internet protocol television (IPTV), one of the most emerging services, offers multimedia streaming services with security, reliability, and relevant quality of service (QoS) / quality of experience (QoE). It provides added values to all the involved players including customers and also brings technical and business challenges to those players. For IPTV services, we expect to adopt the managed network environment for high quality and the Web technologies for personalization to meet the customer's necessity. Web can provide an open, flexible, and agile platform. Therefore, in this paper, we propose personalized IPTV services based on Web-enhanced open platform and present the functional architecture. Technical issues for deploying the proposed services using Web are also provided. The objective of this paper is to analyze the critical architectural and design issues for developing an attractive, high-quality, viable and feasible model for personalised service.
INTRODUCTION
Currently, digital television is gradually replacing analogue TV. Although these digital TV services can be delivered via various broadcast networks (e.g., terrestrial, cable, satellite), Internet Protocol TV (IPTV) over broadband telecommunication networks offers much more than traditional broadcast TV. It can improve the quality that users experience with this linear programming TV service, but it also paves the way for new TV services, such as video-on-demand, time-shifted TV, and network personal video recorder services, because of its integral return channel and the ability to address individual users. IPTV service is considered as the emerging application that has a great potential to generate new revenues for contents and service providers. In the last few years, we have witnessed a rapid growth of multimedia content delivery across the networks. Peer-to-Peer (P2P) networks and user generated content (UGC) are considered as one of the most suitable targeted infrastructure for supporting real time streaming and has played vital role in this growth. One of the major challenges of this approach is to reach the same quality of service of traditional television and commercial IPTV by employing only best effort network layer services. This service gives more choice to the end users to consume TV programs, on-demand content and UGC in a personalized way and beyond any geographical constraints.
ARCHITECTURE OF THE WEB-ENHANCEMENT OF AN IPTV SERVICE
At the beginning of our investigations we have analyzed tens of web-based applications, like youTube, joost, Zattoo, audiTv, videa, mySpace, flickr, hirTV, mtv, DunaTV, etc. We have looked for fancy design ideas, creative features and intuitive navigation with a clear target to create something new. Beside this, we have run speed tests and taken into account the usability, simplicity, platform independency, multiple tools and many features that web-based systems provided. Based on this learning, we started to design our own model. For example, we decided that the clients do not need to download any additional software when they open our web-based application, and there should not be any need for additional installation and configuration. A well-known WebTV application that started with downloadable client software is Joost. After a while they had to abandon the idea of using downloadable player and switched to an all web-based solution. But Joost is not the only one who realised that the future is a web-based solution. Others who had not switch to web-based applications failed or they are about to fail or simply they are not enjoying popularity.
RECOMMENDATION ENGINE
The personalization and recommendation has more and more importance in today’s TV consumption and web behaviour. More than 100 TV channels are a lot more than the consumer can easily manage; furthermore the internet is a huge information labyrinth, so our clear purpose is to make users’ navigation and decision easier. The manually filled profiles have risen seriously, but this is subjective and does not adapt to changing interests of users. There are two frequently used algorithms in today’s recommendation systems: Slope one and Pearson Correlation. These algorithms do not take into consideration the different kind of users and the human nature. A complete intelligent Web personalization system is generally based on Web usage data mining to discover useful knowledge about user access patterns, followed by a recommendation system to act on this knowledge in order to respond to the users’ individual interest. The knowledge discovery component must discover distinct user profiles from Web usage data. Their unsupervised nature also avoids reliance on input parameters or prior knowledge about the number of profiles to be sought. A recent work [11], used item-to-item collaborative filtering as a recommendation strategy in Amazon.com. This approach matches each of the user’s purchased and rated items to similar items then combines those similar items into a recommendation list. A similar-item table is built by finding items which customers trend to purchase together. Unfortunately, they do not present any empirical results, or sufficient details about the proposed technique.
CONCLUSIONS
The IPTV service can deliver TV programs anytime anywhere. IPTV supports both broadcast and unicast services like Live-TV and Video-on-Demand. This paper identified the challenges in delivering web-enhanced IPTV and proposed a framework to provide solutions to those challenges. We have built a prototype of the system and demonstrated its flexible features, integrated EPG, remote controller and recommendation engine. We are firmly convinced that our new EPG design is one of the most competitive designs on the market. We are living in the content-centric world. The user experience of this new media is thought as a key factor for the success of an IPTV service. We are aware that still a lot of innovation is necessary in order to win the battle on “the last millimeters”, namely between the consumer eyes and his brain.