Bradley N. Miller

GroupLens: applying collaborative filtering to Usenet news

newsgroups carry a wide enough spread of messages to make most individuals consider Usenet news to be a high noise information resource. Furthermore, each user values a different set of messages. Both taste and prior knowledge are major factors in evaluating news articles. For example, readers of the rec.humor newsgroup, a group designed for jokes and other humorous postings, value articles based on whether they perceive them to be funny. Readers of technical groups, such as comp.lang.c11 value articles based on interest and usefulness to them—introductory questions and answers may be uninteresting to an expert C11 programmer just as debates over subtle and advanced language features may be useless to the novice. The combination of high volume and personal taste made Usenet news a promising candidate for collaborative filtering. More formally, we determined the potential predictive utility for Usenet news was very high. The GroupLens project started in 1992 and completed a pilot study at two sites to establish the feasibility of using collaborative filtering for Usenet news [8]. Several critical design decisions were made as part of that pilot study, including:

MovieLens unplugged: experiences with an occasionally connected recommender system

Recommender systems have changed the way people shop online. Recommender systems on wireless mobile devices may have the same impact on the way people shop in stores. We present our experience with implementing a recommender system on a PDA that is occasionally connected to the network. This interface helps users of the MovieLens movie recommendation service select movies to rent, buy, or see while away from their computer. The results of a nine month field study show that although there are several challenges to overcome, mobile recommender systems have the potential to provide value to their users today

Using filtering agents to improve prediction quality in the GroupLens research collaborative filtering system

Collaborative filtering systems help address information overload by using the opinions of users in a community to make personal recommendations for documents to each user. Many collaborative filtering systems have few user opinions relative to the large number of documents available. This sparsity problem can reduce the utility of the filtering system by reducing the number of documents for which the system can make recommendations and adversely affecting the quality of recommendations. This paper defines and implements a model for integrating content-based ratings into a collaborative filtering system. The filterbot model allows collaborative filtering systems to address sparsity by tapping the strength of content filtering techniques. We identify and evaluate metrics for assessing the effectiveness of filterbots specifically, and filtering system enhancements in general. Finally, we experimentally validate the filterbot approach by showing that even simple filterbots such as spell checking can increase the utility for users of sparsely populated collaborative filtering systems.

PocketLens: Toward a personal recommender system

Recommender systems using collaborative filtering are a popular technique for reducing information overload and finding products to purchase. One limitation of current recommenders is that they are not portable. They can only run on large computers connected to the Internet. A second limitation is that they require the user to trust the owner of the recommender with personal preference data. Personal recommenders hold the promise of delivering high quality recommendations on palmtop computers, even when disconnected from the Internet. Further, they can protect the users privacy by storing personal information locally, or by sharing it in encrypted form. In this article we present the new PocketLens collaborative filtering algorithm along with five peer-to-peer architectures for finding neighbors. We evaluate the architectures and algorithms in a series of offline experiments. These experiments show that Pocketlens can run on connected servers, on usually connected workstations, or on occasionally connected portable devices, and produce recommendations that are as good as the best published algorithms to date.

MovieLens Unplugged: Experiences with a Recommender System on Four Mobile Devices

Recommender systems have changed the way people shop online. Recommender systems on wireless mobile devices may have the same impact on the way people shop in stores. There are several important challenges that interface designers must overcome on mobile devices: Providing sufficient value to attract prospective wireless users, handling occasionally connected devices, privacy and security, and surmounting the physical limitations of the devices. We present our experience with the implementation of a wireless movie recommender system on a cellphone browser, an AvantGo channel, a wireless PDA, and a voice-only phone interface. These interfaces help MovieLens users select movies to rent, buy, or see while away from their computer. The results of a nine month field study show that although wireless has still not arrived for the majority of users, mobile recommender systems have the potential to provide value to their users today.

Controlling power systems during emergencies: the role of expert systems

Some applications of expert systems to power system control are explored. It is shown how they can be used to alleviate the burden on the operators during emergency periods. To this end, the design and implementation of two applications are presented: an intelligent alarm processor, and a program for the diagnosis of system faults. The special requirements that energy management systems place on expert systems are identified.<<ETX>>

GroupLens for Usenet: Experiences in Applying Collaborative Filtering to a Social Information System

We live in the information overload age. Don’t believe that? Here is some evidence: The world’s total yearly production of print, film, optical, and magnetic content would require roughly 1.5 billion gigabytes of storage. This is the equivalent of 250 megabytes per person for each man, woman, and child on earth. (Lyman and Varian, 2000) The massive amount of content produced each day is changing the way each of us lives our life. Historically, society has coped with the problem of too much information by employing editors, reviewers, and publishers to separate the signal from the noise. The problem is that we do not have enough editors, publishers, and reviewers to keep up with the volume of new content. One solution to this problem is to use technology to allow each of us to act as an editor, publisher, and reviewer for some subset of the rest of society. The technology that enables us to work together to solve the information overload problem for each other is called collaborative filtering.

A zoomable DBMS for brain structure, function and behavior

We have begun a long-term project to build a new kind of database and its enhanced, supporting database management system (DBMS) for international neuroscience research. Because brain research occurs world-wide, our database will be distributed, encouraging rapid, open dissemination of results to a broad audience of neuroscientists. It will conjoin information and experimental results from many disciplines. We envision a zoomable database of the brain tissue itself, in large part embedded in three dimensions (3D), through which one can “fly.” Within this coarse structure, the database will also organize fine-structural, functional and behavioral data. As often as possible, the database will express experimental data in its purest, least analyzed form, so that expensive raw data can be analyzed and reanalyzed by researchers worldwide.

Integrating AI applications in an energy management system

To effectively integrate expert system applications in an energy management system, the authors created an environment that supports all the interfaces between the AI applications and the EMS. This environment also maintains a model of the power system common to all the AI applications. With this environment in place, users can easily plug AI applications into the EMS. The authors have designed and implemented such an environment, which supports three distinct AI applications: intelligent alarm processing, fault diagnosis, and power system restoration. To illustrate the benefits of this approach, they present case studies based on implementations of this environment in three different EMS architectures.

Network requirements for 3-D flying in a zoomable brain database

In laboratories around the world, neuroscientists from diverse disciplines are exploring various aspects of brain structure. Because of the size of the domain, neuroscientists must specialize, making it difficult to fit results together, causing some research efforts to be duplicated because of lack of sharing of information. The authors have begun a long-term project to build a neuroscience research database for brain structure. One aspect of the database is the ability to visualize high-quality, high-resolution micrographs montaged together into 3-D structures as they were in the living brain. As demonstrated in this papers analysis, realistic presentation of these visualizations across computer networks will stress current and proposed gigabit networks. Image compression can reduce network loads, but wide-spread use of the visualizations will still require networks capable of sustaining terabits per second of throughput. >