1. Hybrid Recommender Systems Hybrid recommender systems association two or other recommendation techniques to gain better performance with scarcer of the drawbacks of any individual one. Most commonly, collaborative filtering is combined with some other technique in an attempt to avoid the ramp-up problem. The idea behind hybrid techniques is that a combination of algorithms will provide more accurate and effective recommendations than a single algorithm as the disadvantages of one algorithm can be overcome by another algorithm 1.
1.1. Weighted HybridizationA weighted hybridization is one in which combines the results of different recommended item to generate a recommendation list or prediction by incorporating the scores from each of the techniques in use by a linear combination of recommendation scores. An example of a weighted hybridized recommendation system, the P-Tango system (Claypool et al. 1999) uses such a hybrid 2. The system initially gives collaborative and content-based recommenders equal weight, but gradually gives equal weights at first, and then adjusts the weighting as predictions about user ratings are confirmed or disconfirmed. The benefit of a weighted hybrid is that all the recommender system’s strengths are utilized during the recommendation process in a straightforward way and it is easy to perform post-hoc credit assignment and adjust the hybrid accordingly.
Pazzani’s hybrid recommender does not use numeric scores, but rather treats the output of each recommender (collaborative, content-based and demographic) as a set of votes, which are then combined in a consensus scheme (Pazzani, 1999). 1.2. Switching Hybridization A switching hybrid swaps to one of the recommendation techniques according to a heuristic reflecting the recommender ability to produce a good rating to builds in item-level sensitivity: the system uses some criterion to switch between recommendation techniques; this strategy is that the system is sensitive to the strengths and weaknesses of its constituent recommenders. The switching hybrid has the ability to avoid problems specific to one method e.g. the new user problem of content-based recommender, by switching to a collaborative recommendation system. The Daily Learner system uses a content/collaborative hybrid in which a content-based recommendation method is employed first.
If the content-based system cannot make a recommendation with sufficient confidence, then a collaborative recommendation is attempted.3 This switching hybrid does not completely avoid the ramp-up problem, since both the collaborative and the content-based systems have the “new user” problem. The main disadvantage of switching hybrids is that it usually introduces more complexity to recommendation process because the switching criterion, which normally increases the number of parameters to the recommendation system, has to be determined 3. However, Daily Learner’s hybrid has a “fallback” character – the short-term model is always used first and the other technique only comes into play when that technique fails. Tran & Cohen (1999) proposed a more straightforward switching hybrid. In their system, the agreement between a user’s past ratings and the recommendations of each technique are used to select the technique to employ for the next recommendation. Switching hybrids introduce additional complexity into the recommendation process since the switching criteria must be determined, and this introduces another level of parameterization 4.
However, the benefit is that the system can be sensitive to the strengths and weaknesses of its constituent recommenders. 1.3. Mixed HybridizationWhere it is practical to combine recommendation results of different recommendation techniques simultaneously, instead of having just one recommendation per item, it may be possible to use a “mixed” hybrid, where recommendations from more than one technique are presented together.
Each item has multiple recommendations associated with it. Example of this category, The PTV system (Smyth and Cotter 2000) uses this approach to assemble a recommended program of television viewing schedule for a user by combining recommendations from content-based based on textual descriptions of TV shows and collaborative systems to form a schedule. Profinder (Wasfi, 1999) 5 and PickAFlick(Burke et al. 1997; Burke, 2000) 6, present multiple recommendation sources side-by-side.
Recommendations from the two techniques are combined together in the final suggested program. The mixed hybrid avoids the “new item” start-up problem: the content-based component can be relied on to recommend new shows on the basis of their descriptions even if they have not been rated by anyone. 1.4. Feature Combination Another way to achieve the content/collaborative merger is to treat collaborative information as simply additional feature data associated with each example and use content-based techniques over this augmented data set. For example, Basu, Hirsh & Cohen (1998) report on experiments in which the inductive rule learner Ripper was applied to the task of recommending movies using both user ratings and content features, and achieved significant improvements in precision over a purely collaborative approach. However, this benefit was only achieved by hand filtering content features.
The authors found that employing all of the available content features improved recall but not precision. The feature combination hybrid lets the system consider collaborative data without relying on it exclusively, so it reduces the sensitivity of the system to the number of users who have rated an item. Conversely, it lets the system have information about the inherent similarity of items that are otherwise opaque to a collaborative system 7. The benefit of this technique is that, it does not always exclusively rely on the collaborative data.
1.5. CascadeThe cascade hybridization technique applies an iterative refinement process in constructing an order of preference among different items; the cascade hybrid involves a staged process.
The recommendations of one technique are employed first to produce a coarse ranking of candidates by another recommendation technique from among the candidate set. The hybridization technique is very efficient and tolerant to noise due to the coarse-to-finer nature of the iteration. EntreeC 8 described below, is a cascade hybridization method that used a cascade knowledge-based and collaborative recommender. The recommendations are placed in buckets of equal preference, and the collaborative technique is employed to break ties, further ranking the suggestions in each bucket. Cascading allows the system to avoid employing the second, lower-priority, technique on items that are already well-differentiated by the first or that are sufficiently poorly-rated that they will never be recommended. Because the cascade’s second step focuses only on those items for which additional discrimination is needed, it is more efficient than, for example, a weighted hybrid that applies all of its techniques to all items. In addition, the cascade is by its nature tolerant of noise in the operation of a low-priority technique, since ratings given by the high-priority recommender can only be refined, not overturned. 1.
6. Feature Augmentation The technique makes use of the ratings and other information produced by the previous recommender and that information is then incorporated into the processing of the next recommendation technique. For example, the Libra system (Mooney & Roy 1999) 9 makes content-based recommendation of books on data found in Amazon.com by employing a naïve Bayes text classifier. Feature-augmentation hybrids are superior to feature-combination methods in that they add a small number of features to the primary recommender.
In the text data used by the system is included “related authors” and “related titles” information that Amazon generates using its internal collaborative systems. These features were found to make a significant contribution to the quality of recommendations. The Group Lens research team working with Usenet news filtering also employed feature augmentation (Sarwar et al.
1998). They implemented a set of knowledge-based “filterbots” using specific criteria, such as the number of spelling errors and the size of included messages. These bots contributed ratings to the database of ratings used by the collaborative part of the system, acting as artificial users. With fairly simple agent implementations, they were able to improve email filtering. Augmentation is attractive because it offers a way to improve the performance of a core system, like the Net Perceptions’ Group Lens Recommendation Engine or a naive Bayes text classifier, without modifying it. Additional functionality is added by intermediaries who can use other techniques to augment the data itself. Note that this is different from feature combination in which raw data from different sources is combined. While both the cascade and augmentation techniques sequence two recommenders, with the first recommender having an influence over the second, they are fundamentally quite different.
In an augmentation hybrid, the features used by the second recommender include the output of the first one, such as the ratings contributed by Group Lens’ filter bots. In a cascaded hybrid, the second recommender does not use any output from the first recommender in producing its rankings, but the results of the two recommenders are combined in a prioritized manner.1.7. Meta-level Another way that two recommendation techniques can be combined is by using the internal model generated by one recommendation technique is used as input for another. The model generated is always richer in information when compared to a single rating.
Meta-level 10 hybrids are able to solve the sparsity problem of collaborative filtering techniques by using the entire model learned by the first technique as input for the second technique. Another way that two recommendation techniques can be combined is by using the model generated by one as the input for another. This differs from feature augmentation: in an augmentation hybrid, we use a learned model to generate features for input to a second algorithm; in a meta-level hybrid, the entire model becomes the input. The first meta-level hybrid was the web filtering system Fab (Balabanovic 1997, 1998). In Fab, user-specific selection agents perform content-based filtering using Rocchio’s method (Rocchio 1971) to maintain a term vector model that describes the user’s area of interest.
Collection agents, which garner new pages from the web, use the models from all users in their gathering operations. So, documents are first collected on the basis of their interest to the community as a whole and then distributed to particular users. In addition to the way that user models were shared, Fab was also performing a cascade of collaborative collection and content-based recommendation, although the collaborative step only created a pool of documents and its ranking information was not used by the selection component. A meta-level hybrid that focuses exclusively on recommendation is described by Pazzani (1999) as “collaboration via content”. A content-based model is built by Winnow (Littlestone & Warmuth 1994) for each user describing the features that predict restaurants the user likes. These models, essentially vectors of terms and weights, can then be compared across users to make predictions. More recently, Condliff et al. (1999) have used a two-stage Bayesian mixed-effects scheme: a content-based naive Bayes classifier is built for each user and then the parameters of the classifiers are linked across different users using regression.
LaboUr (Schwab, et al. 2001) uses instance-based learning to create content-based user profiles which are then compared in a collaborative manner 11. The benefit of the meta-level method, especially for the content/collaborative hybrid is that the learned model is a compressed representation of a user’s interest, and a collaborative mechanism that follows can operate on this information-dense representation more easily than on raw rating data. Table 1 Hybridization Methods2. Hybrid Approaches 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 collaborative-based approach to a content-based; or by unifying the approaches into one model 12. 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. The combination of approaches can be done in any of the following ways: separate implementation of algorithms and combining the result, utilizing some content-based filtering in collaborative approach, utilizing some collaborative filtering in content-based approach, creating a unified recommendation system that brings together both approaches.Netflix is a good example of the use of hybrid recommender systems 13. The website makes recommendations by comparing the watching and searching 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).A variety of techniques have been proposed as the basis for recommender systems: collaborative, content-based, knowledge-based, and demographic techniques. Each of these techniques has known shortcomings, such as the well-known cold-start problem for collaborative and content-based systems (what to do with new users with few ratings) and the knowledge engineering bottleneck 14 in knowledge-based approaches.
A hybrid recommender system is one that combines multiple techniques together to achieve some synergy between them.As mentioned earlier, a well-though-out hybridization approach is critical for the success of our two component recommender system. There exist numerous methods to combine collaborative filtering recommender with content-based techniques, but probably not all of them will lead to same prediction Content information forwarded by the contributing recommender should be in an appropriate data format to be processed by the actual collaborative recommender.
Information loss through data transformation should be avoided. Figure 2 Hybrid recommendations System accuracy. We already have discussed several different types of hybrid recommender systems. However, these possible hybrid combinations are not applicable in each situation or for each kind of underlying recommender 15. Due to the fact that collaborative filtering is a well-established way of rating prediction, we want to employ this technique for our actual recommender. In addition, content features of the contributing recommender will support the main recommendation component.
The following figure displays designated hybrid architecture:References1.J.B. Schafer, D. Frankowski, J. Herlocker, S.
Sen Collaborative filtering recommender systems P. Brusilovsky, A. Kobsa, W. Nejdl (Eds.
), The Adaptive Web, LNCS 4321, Springer, Berlin Heidelberg (Germany) (2007), pp. 291-324, 10.1007/978-3-540-72079-9_92. Claypool M, Gokhale A, Miranda T, Murnikov P, Netes D, Sartin M. Combining content- based and collaborative filters in an online newspaper.
In: Proceedings of ACM SIGIR workshop on recommender systems: algorithms and evaluation, Berkeley, California; 2012.3. R. Burke Hybrid recommender systems: survey and experiments User Model User -adapted Interact, 12 (4) (2002), pp. 331-370 Cross Ref View Record in Scopus.4. Billsus D, Pazzani MJ. A hybrid user model for news story classification.
In: Kay J, editor. In: Proceedings of the seventh international conference on user modeling, Banff, Canada. Springer-Verlag, New York; 1999. p.
99–108.5. Wasfi AM. Collecting user access patterns for building user profiles and collaborative filtering. In: Proceedings of the 1999 international conference on intelligent user, Redondo Beach, CA; 1999.
p. 57–64.6. R.
Burke, K. Hammond, B. Young The Find Me approach to assisted browsing IEEE Expert, 12 (4) (1997), pp.
32-40 Cross Ref View Record in Scopus7. Basu C, Hirsh H, Cohen W. Recommendation as classification: using social and content-based information in recommendation.
In: Proceedings of the 15th national conference on artificial intelligence, Madison, WI; 2010. p. 714–20.8. R. Burke Hybrid recommender systems: survey and experiments User Model User-adapted Interact, 12 (4) (2002), pp. 331-370 Cross Ref View Record in Scopus9. R.
J. Mooney, L. Roy Content-based book recommending using learning for text categorization Proceedings of the fifth ACM conference on digital libraries, ACM (2000), pp. 195-204 View Record in Scopus 10. M.J. Pazzani A framework for collaborative, content-based and demographic filtering Artific Intell Rev, 13 (1999), pp. 393-408 No.
5(6) Cross Ref View Record in Scopus11. Schwab I, Kobsa A, Koychev I. Learning user interests through positive examples using content analysis and collaborative filtering. Draft from Fraunhofer Institute for Applied Information Technology, Germany; 2001.12.
Adomavicius, G.; Tuzhilin, A. (June 2005). “Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions”. IEEE Transactions on Knowledge and Data Engineering.
17 (6): 734–749. doi:10.1109/TKDE.2005.99.
13. Gomez-Uribe, Carlos A.; Hunt, Neil (28 December 2015).
“The Netflix Recommender System”. ACM Transactions on Management Information Systems. 6 (4):1–19. doi:10.1145/284394814. Rinke Hoekstra, The Knowledge Reengineering Bottleneck, Semantic Web – Interoperability, Usability, Applicability 1 (2010) 1, IOS Press15. BURKE, ROBIN D.
: Hybrid Web Recommender Systems. In BRUSILOVSKY, PETER, ALFRED KOBSA and WOLFGANG NEJDL (editors): The Adaptive Web, Methods and Strategies of Web Personalization, volume 4321 of Lecture Notes in Computer Science, pages 377–408. Springer, 2007.16. Prof.
Dr.-Ing. habil. Sahin Albayrak Faculty IV – Electrical Engineering and Computer Science Technical University Berlin