Traditional Collaborative Filtering

Traditional collaborative filtering is a type of system used to generate user-specific recommendations based on common items between two similar users. Items most highly ranked by similar users come as the most recommended items. The major problem with this type of system is that, while recommendations are very targeted, they are also incredibly limited as far as what a user will find in his recommendations. This is due to the very limited number of similar users to which one is compared. Problems arise when users have a very narrow pool of items to draw from, and then recommendations that should ideally be similar in liking to a user do not fit that user's taste because recommendations are based on such a small amount of information. Basing recommendations solely on the taste of a small group of users with a common interest makes branching out in exposure to new things and things that are most likely to coincide with taste less and less likely.

Holovaty on the importance of databases

Per Adrian Holovaty, of, among many notable endeavors, EveryBlock, made an appearance in my Digital Media Entrepreneurship class a week or so back and we got to discussing databases and recommendation systems. He told me a bit about a site that he co-founded while in college called Lawrence.com, which is incredibly similar to a project that I'm working on called Collegization.

Basically, both projects aim at getting students into nightlife, while heavily relying on databases of information in order to perform that service, and allowing people to find things that they might like based on location and habits. Both databases include restaurants, venues and events. He told me that the key to running this system, and incorporating a recommendation function into it, is having massive amounts of organized data. Holovaty said that the data, while minute in individual form, is incredibly valuable when amassed for the purposes of Lawrence.com and Collegization. Without these tediously formed databases, it would be impossible for either project to function.

How Internet Cookies Work

I was interested in how sites store information in their databases to remember users and allow them to store preferences. Here's what I found.

This Next

This Next is a site devoted to consumerism. It's main functions are, according to the site, "to explore great product recommendations, get personalized shopping suggestions, and rave about the products you like." Users can see "what's hot" in other major cities, see who is recommending what, and what kind of trends are happening in their own back yards. It also employs a question of the day. For instance today users are asked, "What is your favorite kind of hot chocolate?" This site is very intriguing to me, and I think I'll be exploring it more. The downturn of the economy could put it in a very interesting position, to encourage thrift instead of its perceived devotion to luxury.

Criticker

Criticker is a movie recommendation site based on each user's taste. It employs tags, some simple genre tags like horror, comedy and action, while others are significantly more in-depth: b-movie, robots and sequel being just a fraction of the pages upon pages of ways to define, categorize and describe films. The concept of tagging allows for organization based on likeness. If twenty people tag Coffee & Cigarettes with the word "art," then another film tagged twenty or so times with that same word would be deemed a similar film. This, of course, is simplifying the process and more than one tag would have to match up for films to be recommended, but that is basically how the tagging system allows Criticker users to discover new films.

Helpful articles

I've found a few articles with more detail how exactly some sites take users tastes and make recommendations based on them. First there's What is the Music Genome Project, which details Pandora's time-consuming database work that took years before they could launch, then The New York Times had an article discussing both Pandora and Last.fm. more from the perspective of what users get out of the sites, and finally an interview with Last.fm co-founder Martin Stiksel.

Yelp

Yelp is a site that one can personalize to fit a specific location, for instance, as linked, Tempe, Arizona. Once one creates a profile, upon returning to it Yelp recommends events, restaurants, shopping, nightlife, and an array of other activities and places nearby. The most useful part of this is that these items are ranked by popularity over time, so there are lists of what is currently or recently popular and what has long-term popularity, based on Yelp users' consistently positive reviews. The tagline, "Real people, real reviews," truly encapsulates what the site is about, and what it delivers. My main issue with the site is that things that aren't reviewed on the site will most likely never be noticed.

Competition

I've recently pitched a project idea that revolves around social networking strictly for teenagers. I envision having features that allow for user feedback, with reviews, ratings and recommendations. Based on the feedback I want to be able to tell people overall what's popular, what's actually "good," and calibrate people's tastes. Essentially, I'd like a charting system similar to Last.fm's, that could also conjure recommendations based on other users who share particular likes or dislikes.

High Concept Pitches

DustBuster: Lugging around a heavy, loud, over-sized vacuum cleaner for the smallest of clean-up jobs will no longer be a problem with a cordless, light, hand-held DustBuster with as much power as a vacuum, but a fraction of the size. Easy to use and transport; perfect for anyone needing a quick, thorough cleanup. Target market: Busy moms, bachelors and college kids.

Wite-Out: Wasted paper, illegible copies, letters and documents filled with typos make communication untimely and difficult. Wite-Out makes easy, fast corrections to documents with its quick-drying white liquid solution. Target market: offices, writers, editors, students.

Disposable lighter: A safe, easy and modern alternative to dangerous and environmentally unfriendly stick matches. Carry it in your pocket or purse for a guaranteed quick and convenient light. Target market: smokers, parents, 18-24 year-olds.

Music networking as a function of learning "what's good"

This semester, for my course Digital Media Entrepreneurship, I need to become an expert in some technological aspect. I've chosen music sites that operate by taking in user preferences to determine what people like, and what can be recommended based on other users' preferences. My original idea was to focus on Last.fm, a UK-based site that is my personal favorite for music discovery, widgets and networking based on taste in music. Options within Last.fm are more plentiful than any other site I've come across. One can download a simple program that allows for "scrobbling," or the taking in of data (songs) that are both played in iTunes and on the iPod. Upon scrobbling, Last.fm collects the songs and artists played and creates numerical charts that can be posted as self-updating widgets on virtually any other site with weekly charts tallying up a user's most popular artists.

A topic I'm interested in pursuing is how sites like Last.fm work with others, namely major label dreamboat Rhapsody. An intriguing part of the online music world is that people want all of their listening connected, their charts accounting for all music listened to, so as not to create a bias. This results in many users developing their own ways of connecting the sites. A simple Google search will render results on how to scrobble music from one's personal Rhapsody feed into Last.fm, and do the same with sites like HypeMachine.