Recommender systems or recommendation systems (sometimes replacing "system" with a synonym such as platform or engine) are a subclass of information filtering system that seek to predict the 'rating' or 'preference' that a user would give to an item (such as music, books, or movies) or social element (e.g. people or groups) they had not yet considered, using a model built from the characteristics of an item (content-based approaches) or the user's social environment (collaborative filtering approaches)[1].
Recommender systems have become extremely common in recent years. A few examples of such systems:
Contents |
Recommender systems typically produce a list of recommendations in one of two ways - through collaborative or content-based filtering. Collaborative filtering approaches build a model from a user's past behavior (items previously purchased or selected and/or numerical ratings given to those items) as well as similar decisions made by other users, then use that model to predict items (or ratings for items) that the user may have an interest in [3]. Content-based filtering approaches utilize a series of discrete characteristics of an item in order to recommend additional items with similar properties[4]. These approaches are often combined (see Hybrid Recommender Systems).
The differences between collaborative and content-based filtering can be demonstrated by comparing two popular music recommender systems - Pandora Radio and Last.fm.
Each type of system has its own strengths and weaknesses. In the above example, Last.fm requires a large amount of information on a user in order to make accurate recommendations. This is an example of the cold start problem, and is common in collaborative filtering systems[5]. While Pandora needs very little information to get started, it is far more limited in scope (for example, it can only make recommendations that are similar to the original seed).
When building a model from a user's profile, a distinction is often made between explicit and implicit forms of data collection.
Examples of explicit data collection include the following:
Examples of implicit data collection include the following:
The recommender system compares the collected data to similar and dissimilar data collected from others and calculates a list of recommended items for the user. Several commercial and non-commercial examples are listed in the article on collaborative filtering systems. Recommender systems are a useful alternative to search algorithms since they help users discover items they might not have found by themselves. Interestingly enough, recommender systems are often implemented using search engines indexing non-traditional data.
Montaner provides the first overview of recommender systems, from an intelligent agents perspective.[7] Adomavicius provides a new overview of recommender systems.[8] Herlocker provides an additional overview of evaluation techniques for recommender systems.[9]
One approach to the design of recommender systems that has seen wide use is collaborative filtering. Collaborative filtering methods are based on collecting and analyzing a large amount of information on users’ behaviors, activities or preferences and predicting what users will like based on their similarity to other users[10]. User-based collaborative filtering attempts to model the social process of asking a friend for a recommendation. A key advantage of the collaborative filtering approach is that it does not rely on machine analyzable content and therefore it is capable of accurately recommending complex items such as movies without requiring an "understanding" of the item itself.
One of the most famous examples of Collaborative Filtering is item-to-item collaborative filtering (people who buy x also buy y), an algorithm popularized by Amazon.com's recommender system[11]. Other examples include:
Collaborative filtering approaches often suffer from three problems, cold start, scalability, and sparsity[13].
A particular type of collaborative filtering algorithm uses matrix factorization, a low-rank matrix approximation techique.[14][15]
Another common approach when designing recommender systems is content-based filtering. Content-based filtering methods are based on information about and characteristics of the items that are going to be recommended. In other words, these algorithms try to recommend items that are similar to those that a user liked in the past (or is examining in the present). In particular, various candidate items are compared with items previously rated by the user and the best-matching items are recommended. This approach has its roots in information retrieval and information filtering research.
Basically, these methods use an item profile (i.e., a set of discrete attributes and features) characterizing the item within the system. The system creates a content-based profile of users based on a weighted vector of item features. The weights denote the importance of each feature to the user and can be computed from individually-rated content vectors using a variety of techniques. Simple approaches use the average values of the rated item vector while other sophisticated methods use machine learning techniques such as Bayesian Classifiers, cluster analysis, decision trees, and artificial neural networks in order to estimate the probability that the user is going to like the item.
Direct feedback from a user (usually in the form of a "like" or "dislike" button) can be used to assign higher or lower weights on the importance of certain attributes (using Rocchio Classification or other similar techniques).
As previously detailed, Pandora Radio is a popular example of a content-based recommender system that plays music with similar characteristics to that of a song provided by the user as an initial seed. There are also a large number of content-based recommender systems aimed at providing movie recommendations, a few such examples include Rotten Tomatoes, Internet Movie Database, and Jinni.
Recent research has demonstrated that a hybrid approach, combining collaborative filtering and content-based filtering could be more effective in some cases. Hybrid approaches can be implemented in several ways: by making content-based and collaborative-based predictions separately and then combining them; by adding content-based capabilities to a collaborative-based approach (and vice versa); or by unifying the approaches into one model (see [8] for a complete review of recommender systems). Several studies empirically compare the performance of the hybrid with the pure collaborative and content-based methods and demonstrate that the hybrid methods can provide more accurate recommendations than pure approaches. These methods can also be used to overcome some of the common problems in recommender systems such as cold start and the sparsity problem.
Netflix is a good example of a hybrid system, as it makes recommendations both by comparing the watching habits of similar users (i.e. collaborative filtering) as well as by offering movies that share characteristics with films that a user has rated highly (content-based filtering).
Hundreds of algorithms have been used in the design of recommender systems. The following subsections highlight a few examples.
One of the most commonly used algorithms in recommender systems is the k-nearest neighborhood (k-NN) approach.[16] The k-NN algorithm is a method for classifying objects based on the properties of its closest neighbors in the feature space. In k-NN, an object is classified through a majority vote of its neighbors, with the object being assigned to the class most common amongst its k nearest neighbors (k is a positive integer, typically small). If k = 1, then the object is simply assigned to the class of its nearest neighbor.
The Pearson Correlation is a measure of the correlation (linear dependence) between two variables X and Y, giving a value between +1 and −1 inclusive. In a social network, a particular user's neighborhood with similar taste or interest can be found by calculating the Pearson correlation coefficient. By collecting the preference data of top-N nearest neighbors of a particular user (weighted by similarity), the user's preference can be predicted.
Rocchio Classification is a method of relevance feedback dating back to the 1970's [17]. Rocchio makes use of the Vector Space Model and is based on the assumption that most users have a general conception of which items should be denoted as relevant or non-relevant. User feedback is used to refine a search query by emphasizing or deemphasizing certain terms (similar to how Pandora refines its user recommendations). Through feedback, the user's search query is revised to include an arbitrary percentage of relevant and non-relevant terms as a means of increasing the search engine's recall, and possibly the precision as well. The number of relevant and non-relevant terms allowed to enter a query is dictated by a series of weights in the central equation.
One growing area of research in the area of recommender systems is mobile recommender systems. With the increasing ubiquity of internet-accessing smart phones, it is now possible to offer personalized, context-sensitive recommendations. This is a particularly difficult area of research as mobile data is more complex than recommender systems often have to deal with (it is heterogeneous, noisy, requires spatial and temporal auto-correlation, and has validation and generality problems [18]). Additionally, mobile recommender systems suffer from a transplantation problem - recommendations may not apply in all regions (for instance, it would be unwise to recommend a recipe in an area where all of the ingredients may not be available).
One example of a mobile recommender system is one that offers potentially profitable driving routes for taxi drivers in a city[18]. This system takes as input data in the form of GPS traces of the routes that taxi drivers took while working, which include location (latitude and longitude), time stamps, and operational status (with or without passengers). It them recommends a list of pickup points along a route that will lead to optimal occupancy times and profits. This type of system is obviously location-dependent, and as it must operate on a handheld or embedded device, the computation and energy requirements must remain low.
One of the key events that energized research in recommender systems was the Netflix prize. From 2006 to 2009, Netflix sponsored a competition, offering a grand prize of $1,000,000 to the team that could take an offered dataset of over 100 million movie ratings and return recommendations that were 10% more accurate than those offered by the company's existing recommender system. This competition energized the search for new and more accurate algorithms. On 21 September 2009, the grand prize of US$1,000,000 was given to the BellKor's Pragmatic Chaos team.
The most accurate algorithm in 2007 used an ensemble method of 107 different algorithmic approaches, blended into a single prediction:[19]
Predictive accuracy is substantially improved when blending multiple predictors. Our experience is that most efforts should be concentrated in deriving substantially different approaches, rather than refining a single technique. Consequently, our solution is an ensemble of many methods.
A second contest was planned, but was ultimately canceled in response to an ongoing lawsuit and concerns from the Federal Trade Commission[20].
Building user profiles using collaborative filtering can be problematic from a privacy point of view. Many European countries have a strong culture of data privacy and every attempt to introduce any level of user profiling can result in a negative customer response.[10]
A number of privacy issues arose around the dataset offered by Netflix for the Netflix Prize competition. Although the data sets were anonymized in order to preserve customer privacy, in 2007, two researchers from the University of Texas were able to identify individual users by matching the data sets with film ratings on the Internet Movie Database[21]. As a result, in December 2009, an anonymous Netflix user sued Netflix in Doe v. Netflix, alleging that Netflix had violated U.S. fair trade laws and the Video Privacy Protection Act by releasing the datasets[22]. This led in part to the cancellation of a second Netflix Prize competition in 2010[20].
Much research has been conducted on ongoing privacy issues in this space. Ramakrishnan et.al. have conducted an extensive overview of the trade-offs between personalization and privacy and found that the combination of weak ties (an unexpected connection that provides serendipitous recommendations) and other data sources can be used to uncover identities of users in an anonymized dataset [23].