Personalized Catch-up & DVR: VOD or Linear, That is the Question
by Pancrazio Auteri (Moviri/ContentWise)
The expansion of TV services such as DVR and, more recently, Catch-up have removed the temporal constraint typical of Linear “appointment” TV enabling users to watch content they love at any time and on-demand. However, the DVR and Catch-up TV libraries, while providing a convenient time-shifted “on-demand” consumption, are indeed composed by content previously aired on the Linear TV, so that they have more in common with Linear TV than they have with VOD.
In this talk we will present and discuss the main challenges and some possible solutions to personalize the user experience with content from DVR and Catch-up TV, such as:
- The consumption pattern is strongly affected by the context (e.g., time and device used to access the video service).
- Some content is consumed serially and follows seasonal dynamics (e.g., TV Series).
- The system is fed with a massive and very dynamic stream of data (e.g., new content right after broadcast, signals of user interactions).
- The same piece of content may coexist across multiple services provided by the same operator (e.g., linear schedule, network-DVR, Catch-up TV, subscription VOD, rental VOD).
About the Speaker
Pancrazio is CTO at ContentWise, focusing on personalization and analytics. Before joining ContentWise, Pancrazio was at TiVo as director of product management for VOD, metadata, TV Everywhere and third-party content apps. With a background of intranet design, digital television and analytics, he spent the last decade at the intersection of Internet and TV, leading the Italian team that in 2007 introduced the first digital terrestrial Internet-DVR featuring HTML apps. In 2011 Pancrazio moved to California to drive the multi-screen and big data evolution of a leading IPTV platform. In his spare time you can find him hacking local ingredients to cook Sicilian food.
Slides
http://www.slideshare.net/PancrazioAuteri/complicated-tv-made-easy-again
Recommendations for Live TV
by Jan Neumann & Hassan Sayyadi (Comcast)
Despite the rise in video-on-demand consumption, live TV is still the most popular way to consume video entertainment. At Comcast we are developing novel ways to make it easy for our customers to access the live TV content that is interesting and relevant for them at the current moment. In this talk, we will describe some of the latest research at Comcast Labs on learning the favorite stations and programs for a customer at a given time of day, personalizing their TV guide, and informing our customers of what is trending on TV and social media at that moment, so that they can participate in the shared experience of live TV. We will explain how usage data is processed using both batch and real-time approaches to personalize the experience for Comcast’s customers.
About the Speakers
Jan Neumann is a senior manager at Comcast Labs Washington DC where he leads the big data and multi-media content analysis research teams. Before Comcast, he worked for Siemens Corporate Research on various computer vision related projects. He holds a Ph.D. in Computer Science from the University of Maryland, College Park.
Hassan Sayyadi is a Lead Researcher at the Comcast Research Lab in Washington DC. His research interests lie in the general field of machine learning with a special focus on recommendation and ranking problems in the TV and video domain. At Comcast, he is leading the personalization research team, that develops novel personalized search, browse, and recommendation solutions to improve the experience for Comcast’s customers. He holds a Ph.D. in Computer Science from the University of Maryland, College Park.
The Application of Recommender Systems in a Multi Site, Multi Domain Environment
by Steven Bourke (Schibsted)
Schibsted is one of the largest media companies in the world; it has more than 200 million unique users a month, split across 39 countries across the world. In this talk we will focus on our recommendation efforts both in online news, and marketplaces. Specifically we look into the commonalities between both domains. Additionally, we also look into some of the unique problems from each domain and how we attempt to solve them. The main themes will include dealing with time sensitive recommendations and as well as dealing with scarcity in recommendations.
About the Speaker
Steven Bourke is a data scientist at Schibsted. He is a member of the recently formed Schibsted product and technology group where he works on building out and researching personalisation techniques to be used across Schibsted’s online properties. Previous to that he performed research in the field of recommender systems at the University of Dublin, under the supervision of Prof. Barry Smyth.
We Know Where You Should Work Next Summer: Job Recommendations
by Fabian Abel (XING)
Business-oriented social networks like LinkedIn or XING support people in discovering career opportunities. In this talk, we will focus on the problem of recommending job offers to millions of XING users. We will discuss challenges of building a job recommendation system that has to satisfy the demands of both job seekers who have certain wishes concerning their next career step and recruiters who aim to hire the most appropriate candidate for a job. Based on insights gained from a large-scale analysis of usage data and profile data such as curriculum vitae, we will study features of the recommendation algorithms that aim to solve the problem.
About the Speaker
Fabian is a Data Science team lead at XING. He enjoys working on large-scale data mining problems and delivering data products that do something meaningful. Before he joined XING in 2012, he was working as a postdoc at TU Delft, the Netherlands, and as PhD student at L3S Research Center in Hanover, Germany, researching user behavior and personalized information retrieval on the Social Web.
Slides
http://www.slideshare.net/fobabel/we-know-where-you-should-work-next-summer-job-recommendations
Assessing Expertise in the Enterprise: The Recommender Point of View
by Aleksandra (Saška) Mojsilović (IBM Research)
Some of the largest worldwide employers today are knowledge-based enterprises whose most important asset is human capital. Knowledge workers are unique, each having individualized skills, competencies and expertise, which constantly evolve and expand. Managing and planning for such a workforce critically depends on the ability to construct complete, accurate, and real-time representation and inventory of the expertise of employees in a form that integrates with business processes. In this session Saška will describe how enterprise expertise assessment process can be posed as predictive modeling and recommendation problem, and will present results and lessons learned from an actual deployment of IBM Expertise, a corporate-wide expertise recommendation and management system.
About the Speaker
Aleksandra (Saška) Mojsilović manages Data Science Group at the IBM T. J. Watson Research Center in Yorktown Heights, New York. Saška is one of the pioneers of business analytics at IBM an in the industry; throughout her career she championed innovative uses of analytics for business decision support: from the early identification of client risk via predictive modeling, to the estimation of outsourcing benefits via signal analysis in support of IBM marketing campaigns, retention analytics, and identifying and recommending experts in the enterprise. For her technical contributions and the business impact of her work, Saška was appointed an IBM Fellow, the company’s highest technical honor. Saška received her PhD in Electrical Engineering in 1997 from the University of Belgrade, Belgrade, Serbia. She has worked at Bell Laboratories (1998-2000) and IBM Research (2000-present). Her main research interests include multidimensional signal processing, pattern recognition and machine learning, with applications to business analytics, healthcare, financial modeling, multimedia and social systems. She is the author of over 100 publications and holds 14 patents. Saška received a number of awards for her work, including the IEEE Young Author Best Paper Award, INFORMS Wagner Prize, IEEE International Conference on Service Operations and Logistics and Informatics Best Paper Award, European Conference on Computer Vision Best Paper Award, IBM Gerstner Award, IBM Market Intelligence Award and several IBM Outstanding Technical Achievement Awards.
