Within the Web, the annotation of content has become a common way to provide e cient navigation and recommendation of resources. In the future, TV sets with integrated Web capabilities will o er tagging as a tool for content organization in the realm of home entertainment. The recommendation of TV content is a challenging task as a system has to consider each user's individual preferences without getting too speci c. We present a strategy which employs user-generated tags in a exible way to address this issue. Our approach provides two di erent ways of semantic ranking for TV program lists: The rst allows a higher ranking of programs that t well to the user's personal likings. The second introduces collaborative aspects and therefore promotes a community-driven approach rather than an individual way of recommendation.
2.
In recent years, the fusion of television and the Web has already begun. In this context, the integration of content from the Web into television and vice versa are two important and not yet completed tasks. Considering the characteristics of both information sources, the following turns out: while television is consumed mostly passively, Web content usually o ers a high degree of user interaction. However, in the next years, this distinction will become more and more blurred. In particular, television will o er common ways of interaction that are currently only well known from the Web, especially social annotation of content.
Our approach applies user-generated tags in order to provide recommendation of TV content. As a result of an information ltering process, we provide two rankings of a program list, each of which is based on the same data but employs di erent ways of user modeling. The personalized ranking focuses on the semantic similarity between 2.1
As input for the recommendation approach, we consider two entities: The rst is an upcoming program, which has not been tagged yet. The second is a user pro le that contains the history of previously watched programs along with the tags assigned by the users.
Finding tags for an upcoming program, which is a
candidate for recommendation, is a non-trivial task as users
commonly assign tags after and not before media
consumption. There exist various options to tackle this problem:
Keywords can be extracted from the program descriptions
(as it is done by tvister1) and reused as tags. Furthermore, a
1tvister - http://www.tvister.de/
professional team could tag upcoming programs in advance.
It is clear that the creation of tag clouds in both of these
ways di ers from the dynamic process of community
tagging. In [
For the personalized and the collaborative part of
recommendation, we use two di erent representations of the same
user pro le (containing the tagging history). To accentuate
this, we refer to table 2, which shows a small user pro le
that is present in our dataset [
The comparison of upcoming programs with the personalized version of the user pro le and also with the collaborative one, results in two di erent program rankings.
In the following, we refer to the user pro le as either the personalized or the collaborative representation. 2.2
A similarity measure is used to compare the tag clouds of
the programs in the user pro le to the ones of the upcoming
programs. We represent tag clouds as vectors in a similar
way as items are represented as vectors of user ratings in the
collaborative ltering domain [
action (3.937), polizei (1.999), krimi (1.995), spannend (1.990), autos (0.989), aufregend (0.898), aktion (0.896), serie (0.879)
lm (1.617), comic (1.613), geschichte (1.597), spiel lm (0.859), spass (0.859), comik (0.859), lustig (0.859), zeichentrick (0.858)
zeichentrick (6.357), homer (3.653), comedy (3.552), kult
(3.535), lustig (3.434), humor (2.529), serie (1.957),
cartoon (1.773), simpsons (1.711), entspannung (1.549),
amerika (1.498), fun (0.761), marge (0.899), james brooks
(0.824), chillen (0.750), neue folgen (0.736), bart (0.726)
[
P (wp;t t2Tpq wp)(wq;t
wq)
P (wp;t t2Tpq wp)2 P (wq;t t2Tpq wq)2
1 + 1CC
A The value wk stands for the average tag weight of the program k: 1 X ~ jkj t2Tk wk;t Note that the resulting similarity score lies between 0 and 1 with a neutral point at 0.5 and equality at 1. 2.3
After having measured the similarity of the new program to all programs in the user pro le, we need to aggregate these scores to a nal one for each upcoming program. It does not make sense to aggregate the similarity scores of all programs in the pro le as users do often like more than one program genre. Hence, we only aggregate the similarity scores of the k-nearest neighbors (k-NN) in order to aim at speci c genres the user prefers. This helps to obtain more accurate results as inter-genre measurements often result in a neutral similarity score that would be incorporated in every aggregation. For the aggregation of the scores of the k-NN, we use a weighted average approach with the scores being the weight (what results in squaring the similarity scores). For the programs within the k-nearest neighbors kN N P rof ile and the upcoming program pnew this leads to the following formula: of the recommender system highly relies on the user's taste and therefore implements her individual preferences.
For a single top-N listing, it is possible to linearly combine both types of scores for each program. agg(pnew) =
P p2kNN The aggregated score of an upcoming program can be interpreted as the user's degree of preference for it. In our approach, this value is used to provide a ranking within the list of upcoming programs. 3.
By using the upcoming programs of table 1 and the prole of table 2 as input data, we do now exemplify how the aforementioned pro le representations can provide di erent scores and rankings. It needs to be pointed out that, for the reasons of clarity and brevity, the chosen user pro le is very small (only seven tagged programs) and also the short list of upcoming programs does not relate to a real case scenario (usually more than 200 concurrent programs).
The personalized as well as the collaborative rankings,
shown in table 3, demonstrate that a top-N recommendation
is possible with only few ratings. By considering tags,
similarities between TV programs can be determined although
they are not strongly correlated through content or
metadata. In our case, the Asterix movie nearly gets the same
score (on both sides) as the upcoming Simpsons episode
although it has no direct correlation (through TV metadata or
content) to one of the user's previously watched programs.
In contrast, the upcoming Simpson episode does have this
link: the user has already watched two episodes before.
Therefore the reasonably high score of the Asterix movie
indicates that, even with a small pro le, the use of tags as
semantic descriptors might help to overcome the common
problem of overspecialization. This also underlines our
efforts to provide collaborative semantic tag prediction [
It is apparent that the ranking of the three programs in table 3 is the same for the personalized and for the collaborative representation of the user pro le. However, as the di erences between the scores in both lists indicate, the ranking would strongly di er taken a larger and more realistic number of upcoming programs (> 200) into account. The similarity scores of the collaborative ranking highlight the community factor of the ranking. The personalized part 4.
This paper presents two feasible and promising approaches to provide top-N recommendations through collaborative tagging. Moreover, it is demonstrated that the utilization of user-generated tags might help to overcome the problem of overspecialization in the emergent domain of TV recommendation.
For the future work, we plan to conduct a thorough evaluation of the proposed approach that also includes a user survey. Furthermore, the similarity measurements can be enhanced through lemmatization of tags in combination with ontology matchings between tag clouds.