Preparing the stand for the OMORE presentation

1. Movie cross-references: It is not known what movie of one provider corresponds to what other movie page of another provider. Providers may be commercial pages like Amazon.com, review pages like RottenTomatoes.com or knowledge bases such as IMDb.com
2. Retrieval of movie cross-references: No flexible search service exists to retrieve movie cross-references based on movie title and release year information.
3. Maintenance of movie cross-references: A vast amount of potential movie cross-references exists that is difficult to gathered with a Web crawler approach. In addition, the set of movie pages increases fast.

We came up with the following solutions:

1. Movie cross-references: A knowledge base is needed to persist movie cross-references. Concretely, for all movies the information of (1) what movie pages represent the movie as its content and (2) what movie pages represent the commercial product of a movie such as DVD, Blu-Ray, VHS or even Video-On-Demand (VoD). This semantical distinction has to be done, because a movie represented in VHS and Blu-Ray are not the same but show the same movie. Therefore, we decided to apply Semantic Web technologies to persist movie cross-references. We applied D2R that maps data from a relational database management system to RDF. RDF is the basic format to represent resources semantically. Our knowledge base of movie cross-references is called LiMo.
2. Retrieval of movie cross-references: A search service has to be provided, that is able to provide even fuzzy search on the movie cross-references knowledge base. We reason for fuzzy search because the movie are presented quite heterogeneously among different Web pages. Movie titles may be misspelled, transformed or even extended in various way. Especially on online shops, we experienced that the movie titles are extended with information about many variants of special or collector’s edition and the type of medium the movie is provided. Instead of trying to extract the original title from the unpurified title, we decided to apply fuzzy search over movie titles and release year to retrieve movies. Our movie retrieval service can be accessed at MOLookup.
3. Maintenance of movie cross-references: A Web crawler approach is not feasible due to the time latency and the need for resources. Thus, we decided to invent a collaborative approach. Whenever a user browses a new movie Web page that is not yet cross-referenced with LiMo, OMORE automatically uses the movie retrieval services MOLookup and provides the current URL of the new movie page. Then, this URL is cross-referenced to the retrieved movie. With that approach we automatically gather all the relevant movie cross-references with the user’s help. This way, normal users even contribute to the Semantic Web without knowing it.

With this new approach, we decided to participate in the this year’s Semantic Web Challenge that is co-located with the International Semantic Web Conference 2009 (ISWC) in Washington D.C. We were 16 participants that made it to the Semantic Web Challenge in Washington D.C. We presented our movie recommender system and its revised architecture besides the official Poster and Demonstrations session the main conference. Our secret weapon to attract many people to our stand was Swiss chocolate. And well, it worked out . The official time for the challenge presentation was 19:15-21:15. But people already showed up to our stand at half past 6 and kept coming by until 10 in the evening. One reason my be of course the chocolate , but also the viral marketing that people started that saw our challenge. Overall, people were really excited about our personal and private movie recommender system that even provides cross-site movie recommendations.
Despite the great success, we didn’t made it to finals. However, the challenge was a really nice experience and we had still have the great success having people excited about our OMORE.

Abstract

Online stores and Web portals bring information about a myriad of items such as books, CDs, restaurants or movies at the user’s fingertips. Although, the Web reduces the barrier to the information, the user is overwhelmed by the number of available items. Therefore, recommender systems aim to guide the user to relevant items. Current recommender systems store user ratings on the server side. This way the scope of the recommendations is limited to this server only. In addition, the user entrusts the operator of the server with valuable information about his preferences.
Thus, we introduce the private, personal movie recommender OMORE, which learns the user model based on the user’s movie ratings. To preserve privacy, OMORE is implemented as Firefox add-on which stores the user ratings and the learned user model locally at the client side. Although OMORE uses the features from the movie pages on the IMDb site, it is not restricted to IMDb only. To enable cross-referencing between various movie sites such as IMDb, Amazon.com, Blockbuster, Netflix, Jinni, or Rotten Tomatoes we introduce the movie cross-reference database LiMo which contributes to the Linked Data cloud.

Presentation

In the following, you can watch my presentation I prepared for the Semantic Web Challenge:

Poster

In the following, you can see a preview of the Semantic Web Challenge 2009 poster titled “OMORE”:

Poster presentation of OMORE at the Semantic Web Challenge 2009

Downloads

We include the papers on this page to ensure timely dissemination on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by the copyrights. These works may not be reposted without the explicit permission of the copyright holder.

• PDF
• RDF
• BibTex

The Research day starts at 16.15 at the Department of Informatics. My adviser Prof. Abraham Bernstein opens the session with his talk on “Dem Gehirn beim Denken zusehen – Wie die Informatik neue Welten erschliesst”. Afterwards, all people present their current work on posters. Everyone will be open for questions and discussions.

I’ll participate and will present two different posters. One poster is about a distributed collaborative recommender system approach. My second poster is about a , OMORE, a Firefox Add-on that enables cross-site recommendation for movies.

You may find additional informations about the upcoming Research Day. Here you get important information how to reach us.

Distributed Collaborative Recommender System

OMORE - Firefox Add-on for cross-site recommendations

OMORE recognizes a Movie Web site and adds automatically rating and recommendation functionality to a Movie Web site at Amazon.com

Thus, we need a recommender system that (1) recognizes items over various Web pages as the same and remembers the ratings for those, (2) applies algorithms to provide recommendations and (3) smoothly integrates the rating and recommendation functionality directly in the Web site. We found the basic infrastructure for such a recommender system in the Firefox Add-on API and WEKA, a common data mining library.
We formulated these requirements as a master thesis. We were very happy to engage Tobias, an excellent master student.

The described recommender system implemented as Firefox Add-on can be downloaded at s.e.a.l. group site.

Abstract

Online stores and Web portals bring information about a myriad of items such as books, CDs, restaurants or movies at the user’s fingertips. Although, the Web reduces the barrier to the information, the user is overwhelmed by the number of available items. Therefore, recommender systems aim to guide the user to relevant items. Current recommender systems store user ratings on the server side. This way the scope of the recommendations is limited to this server only. In addition, the user entrusts the operator of the server with valuable information about his preferences.

In this thesis, we introduce our recommender system OMORE, a private, personal movie recommender, which learns the user model based on the user’s movie ratings. To preserve privacy, OMORE is implemented as a Mozilla Firefox add-on, which stores the user’s ratings and the learned user model locally at the client side. Although OMORE makes use of the movie features, which are provided by the different movie pages on the Amazon.com, Blockbuster, Netflix and Rotten Tomatoes.

Zusammenfassung

Online-Geschäfte und Web Portale bieten einem Kunden im Allgemeinen eine riesige Auswahl an Filmen oder Büchern an. Oftmals ist dieser aber mit der riesigen Auswahl an vorhandenen Artikeln überfordert und braucht Unterstützung, um die für ihn interessanten Produkte auch wirklich zu finden. Empfehlungssysteme haben sich bewährt und sind sehr erfolgreich beim Filtern von grossen Datenbeständen. Doch nur wenige Portale wie das Online-Geschäftshaus Amazon bieten einem Kunden ein solches Empfehlungssystem zur aktiven Unterstützung an.

In der Regel basieren die von einem Empfehlungssystem vorgeschlagenen Produktempfehlungen auf den Bewertungen von anderen Kunden. Diese werden in den heutzutage verfügbaren Empfehlungssystemen häufig von den Anbietern eines Web Portals individuell verwaltet, so dass sie dadurch nicht auf anderen Portalen wie zum Beispiel dem Online DVD Verleih Netflix verwendet werden können. Zudem vertraut ein Kunde einem Anbieter eines solchen Portals oft sehr vertrauliche Informationen über sein Kaufverhalten und seine Präferenzen an.

In dieser Arbeit soll daher ein portalunabhängiges Empfehlungssystem entwickelt werden, welches direkt im Web-Browser integriert ist. Das von uns auf den Namen OMORE getaufte Empfehlungssystem, ist ein auf Sicherheit ausgerichtetes personalisiertes Empfehlungssystem für Filmliebhaber, welches als Erweiterung für den Mozilla Firefox Browser angeboten wird. Es lernt die Benutzerpräferenzen basierend auf den Filmbewertungen eines Benutzers und speichert das gelernte Model der Benutzerpräferenzen lokal auf dem System des Benutzers ab. Dadurch wird sichergestellt, dass die Benutzerpräferenzen vor unbefugtem Zugriff geschützt sind. OMORE bietet einem Benutzer portalübergreifende Empfehlungen an, wobei die aktuelle Implementierung die Filmseiten von Amazon.com, Blockbuster, der Internet Movie Database, Netflix und Rotten Tomatoes umfasst.

Tobias Bannwart: “OMORE – Private, Personal Movie Recommendations implemented in a Mozilla Firefox Add-on“, ed. by Amancio Bouza, Gerald Reif and Harald C. Gall, University of Zurich, July 2009. (master thesis)

Downloads:

• OMORE Firefox Add-on

Participants of the IRMLeS workshop

Our current work on a probabilistic approach to compute partial user preference similarities was accepted and published at the 1st International Workshop on Inductive Reasoning and Machine Learning for the Semantic Web (IRMLeS) at the 6th European Semantic Web Conference (ESWC) 2009. The paper is available online at CEUR-WS.org as volume 474. The presentation is available at the IRMLeS Web page.

The general idea is that people may share similar preferences only partially. For instance, a person may like Italian food like another person but not Chinese food. But the other person does like Chinese food. Traditional collaborative filtering computes global preference similarity and fail detect this relation. Our approach computes is able to compute partial preference similarities on the basis of hypothesized user preferences. The hypothesized user preferences are learned applying traditional machine learning algorithms. We could show, that our approach performs significantly better then traditional user-based collaborative filtering. Especially in cases where people have only few common rated items. The strength of our approach are the use of partial preference similarities and using hypothesized user preferences instead of item ratings that are always biased.

It was my first real presentation at a conference and it was a great success. I got very positive feedback on it. But I also noticed that a too fancy presentation may irritate some people. Well, I just had the new version of Keynote installed on my Mac and thus, I had to try out the new fancy features. This workshop was one of the most successful at the this year’s ESWC measured by number of participants. I also enjoyed the workshop dinner where I participated interesting discussion on artificial intelligence, data mining in practice, football and Shakespeare.

At the conference, I got in touch with some very interesting people. Especially at the very well organised poster session and after the conference dinner. Unfortunately, it was the last European Semantic Web Conference because the organizers decided to have the industry as main target. Thus, the abbreviation of ESWC stands now for the Extended Semantic Web Conference.

Abstract

Recommender systems play an important role in supporting people getting items they like. One type of recommender systems is user-based collaborative filtering. The fundamental assumption of user-based collaborative filtering is that people who share similar preferences for common items behave similar in the future. The similarity of user preferences is computed globally on common rated items such that partial preference similarities might be missed. Consequently, valuable ratings of partially similar users are ignored. Furthermore, two users may even have similar preferences but the set of common rated items is too small to infer preference similarity. We propose first, an approach that computes user preference similarities based on learned user preference models and second, we propose a method to compute partial user preference similarities based on partial user model similarities. For users with few common rated items, we show that user similarity based on preferences significantly outperforms user similarity based on common rated items.

Presentation

In the following, you can watch my paper presentation I gave at the IRMLeS workshop:

Downloads

We include the papers on this page to ensure timely dissemination on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by the copyrights. These works may not be reposted without the explicit permission of the copyright holder.

• PDF, alternatively from the Zurich Open Repository and Archive, DDIS group or s.e.a.l. group.
• BibTex
• Presentation slides

Amancio Bouza at the poster presentation of SemTree

Our current work on a ontology-based decision tree algorithm to learn user preferences was accepted and published at 7th International Semantic Web Conference (ISWC) 2008 in Karlsruhe (Germany). The paper is available online at CEUR-WS.org as volume 401.

It was my first participation of a conference where I had my first opportunity to present our early work on a novel approach on inducing a decision tree based on a domain ontology. SemTree is a machine learning algorithm based on the concept of decision tree induction. The novelty about SemTree is the usage of a domain ontology to learn more accurate models. That has been shown in a preliminary evaluation on a movie dataset.
At the poster session I had great conversations and discussions about recommender systems and my approach. But the disadvantage of presenting a poster to the research community is that oneself cannot take a look after the other interesting posters and demos.

To my surprise, Karlsruhe is a very nice german city dominated by students – a typical students town. That shows the high frequency of small bars and restaurants. The castle looks pretty and the very large castle park is absolutely beautiful. I would have liked to enjoy this park in the summer and spend my leisure time with friends there.

Abstract

Recommender systems play an important role in supporting people when choosing items from an overwhelming huge number of choices. So far, no recommender system makes use of domain knowledge. We are modeling user preferences with a machine learning approach to recommend people items by predicting the item ratings. Specifically, we propose SemTree, an ontology-based decision tree learner, that uses a reasoner and an ontology to semantically generalize item features to improve the effectiveness of the decision tree built. We show that SemTree outperforms comparable approaches in recommending more accurate recommendations considering domain knowledge.

Presentation

In the following, you can see a preview of the SemTree poster titled “seMANtics IN TREEs” in allusion to the american TV series “Man in Trees”:

Poster presentation of SemTree at the ISWC 2008

Downloads

We include the papers on this page to ensure timely dissemination on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by the copyrights. These works may not be reposted without the explicit permission of the copyright holder.

• PDF
• RDF
• BibTex