In many commercial systems, the 'best bet' recommendations are shown, but the predicted rating values are not. This is usually referred to as a top-N recommendation task, where the goal of the recommender system is to find a few specific items which are supposed to be most appealing to the user. Common methodologies based on error metrics (such as RMSE) are not a natural fit for evaluating the top-N recommendation task. Rather, top-N performance can be directly measured by alternative methodologies based on accuracy metrics (such as precision/recall).An extensive evaluation of several state-of-the art recommender algorithms suggests that algorithms optimized for minimizing RMSE do not necessarily perform as expected in terms of top-N recommendation task. Results show that improvements in RMSE often do not translate into accuracy improvements. In particular, a naive non-personalized algorithm can outperform some common recommendation approaches and almost match the accuracy of sophisticated algorithms. Another finding is that the very few top popular items can skew the top-N performance. The analysis points out that when evaluating a recommender algorithm on the top-N recommendation task, the test set should be chosen carefully in order to not bias accuracy metrics towards non-personalized solutions. Finally, we offer practitioners new variants of two collaborative filtering algorithms that, regardless of their RMSE, significantly outperform other recommender algorithms in pursuing the top-N recommendation task, with offering additional practical advantages. This comes at surprise given the simplicity of these two methods.

Performance of recommender algorithms on top-n recommendation tasks

John Seely Brown is a Visiting Scholar and Advisor to the Provost at the University of Southern California (USC) and Independent Co-Chairman of a New Deloitte Research Center. He is the former Chief Scientist of Xerox and Director of its Palo Alto Research Center (PARC). Many of his publications and presentations are on his website (http://www.johnseelybrown.com). Richard P. Adler is a Research Affiliate at the Institute for the Future in Palo Alto and Principal of People & Technology, a research and consulting firm in Cupertino, California. More than one-third of the world’s population is under 20. There are over 30 million people today qualified to enter a university who have no place to go. During the next decade, this 30 million will grow to 100 million. To meet this staggering demand, a major university needs to be created each week. —Sir John Daniel, 1996 T he world has become increasingly “flat,” as Tom Friedman has shown. Thanks to massive improvements in communications and transportation, virtually any place on earth can be connected to markets anywhere else on earth and can become globally competitive.1 But at the same time that the world has become flatter, it has also become “spikier”: the places that are globally competitive are those that have robust local ecosystems of resources supporting innovation and productiveness.2 A key part of any such ecosystem is a well-educated workforce with the requisite competitive skills. And in a rapidly changing world, these ecosystems must not only supply this workforce but also provide support for continuous learning and for the ongoing creation of new ideas and skills. Minds on

/pdf/minds-on-fire-open-education-the-long-tail-and-learning-2-0-2qnq4fh0by.pdf

Minds on Fire: Open Education, the Long Tail, and Learning 2.0

As social network sites like MySpace and Facebook emerged, American teenagers began adopting them as spaces to mark identity and socialize with peers. Teens leveraged these sites for a wide array of everyday social practices - gossiping, flirting, joking around, sharing information, and simply hanging out. While social network sites were predominantly used by teens as a peer-based social outlet, the unchartered nature of these sites generated fear among adults. This dissertation documents my 2.5-year ethnographic study of American teens' engagement with social network sites and the ways in which their participation supported and complicated three practices - self-presentation, peer sociality, and negotiating adult society. My analysis centers on how social network sites can be understood as networked publics which are simultaneously (1) the space constructed through networked technologies and (2) the imagined community that emerges as a result of the intersection of people, technology, and practice. Networked publics support many of the same practices as unmediated publics, but their structural differences often inflect practices in unique ways. Four properties - persistence, searchability, replicability, and scalability - and three dynamics - invisible audiences, collapsed contexts, and the blurring of public and private - are examined and woven throughout the discussion. While teenagers primarily leverage social network sites to engage in common practices, the properties of these sites configured their practices and teens were forced to contend with the resultant dynamics. Often, in doing so, they reworked the technology for their purposes. As teenagers learned to navigate social network sites, they developed potent strategies for managing the complexities of and social awkwardness incurred by these sites. Their strategies reveal how new forms of social media are incorporated into everyday life, complicating some practices and reinforcing others. New technologies reshape public life, but teens' engagement also reconfigures the technology itself.

/pdf/taken-out-of-context-american-teen-sociality-in-networked-1en7i1011h.pdf

Taken Out of Context: American Teen Sociality in Networked Publics

Although an extensive body of research has emerged on marketing in computer-mediated environments (CMEs), the literature remains fragmented. As a result, insights and findings have accumulated without an overarching framework that provides structure and guidance to the rapidly-growing literature. We believe this is detrimental to long-term knowledge development in this area. To address this issue, we organize and synthesize findings from the literature using a framework structured around four key interactions in CMEs: consumer-firm interactions, firm-consumer interactions, consumer-consumer interactions, and firm-firm interactions. The proposed framework serves a valuable organizational function and helps identify a broad spectrum of gaps in the literature to advance the next generation of knowledge development.

Marketing in Computer-Mediated Environments: Research Synthesis and New Directions

Generating alternative queries, also known as query suggestion, has long been proved useful to help a user explore and express his information need. In many scenarios, such suggestions can be generated from a large scale graph of queries and other accessory information, such as the clickthrough. However, how to generate suggestions while ensuring their semantic consistency with the original query remains a challenging problem.In this work, we propose a novel query suggestion algorithm based on ranking queries with the hitting time on a large scale bipartite graph. Without involvement of twisted heuristics or heavy tuning of parameters, this method clearly captures the semantic consistency between the suggested query and the original query. Empirical experiments on a large scale query log of a commercial search engine and a scientific literature collection show that hitting time is effective to generate semantically consistent query suggestions. The proposed algorithm and its variations can successfully boost long tail queries, accommodating personalized query suggestion, as well as finding related authors in research.

/pdf/query-suggestion-using-hitting-time-4g0x31db6w.pdf

Query suggestion using hitting time

The Long Tail: Why the Future of Business Is Selling Less of More

Television Is the New Television: The Unexpected Triumph of Old Media in the Digital Age

/pdf/ici-etait-radio-canada-14ddtzjok8.pdf

Ici était Radio-Canada

/pdf/construire-la-nation-au-petit-ecran-le-canada-une-histoire-4c0hca1tu2.pdf

Robert Armstrong

Papers

The Long Tail: Why the Future of Business Is Selling Less of More

Television Is the New Television: The Unexpected Triumph of Old Media in the Digital Age

Ici était Radio-Canada

Construire la nation au petit écran : Le Canada, une histoire populaire de CBC / Radio-Canada (1995 – 2002).

Les barbares numériques: résister à l’invasion des GAFAM. Par Alain Saulnier