MD
M.J.C. Dingjan
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
Recommender systems focus on automatically surfacing suitable items for users from digital collections that are too large for the user to oversee themselves. A considerable body of work exists on surfacing items that match what a user liked in the past; this way, the recommender system will exploit its knowledge of a user's comfort zone. However, application scenarios exist in which it is explicitly important to offer the user opportunities to explore items beyond their existing comfort zones. In such cases, the recommender should include items for which there is less existing evidence that the user will like them. This calls for the recommender to explore to what extent a user would be tolerant to exploration. In this thesis, we consider that different users will likely have different preferences with respect to item exploration. We propose personalized item filtering techniques for this, modeled under a multi-armed bandit framework, that consider (1) how and where exploration vs. exploitation items should be distributed in a result overview and (2) how adventurous exploration items can be, considering a user's general willingness to explore and existing familiarity with item categories in the collection. We present the results of a survey and an online quantitative experiment with 43 users of Muziekweb, a public music library that encourages taste broadening in the Netherlands, demonstrating the effectiveness of both our proposed filtering techniques that aim for personalized exploration.
...
Recommender systems focus on automatically surfacing suitable items for users from digital collections that are too large for the user to oversee themselves. A considerable body of work exists on surfacing items that match what a user liked in the past; this way, the recommender system will exploit its knowledge of a user's comfort zone. However, application scenarios exist in which it is explicitly important to offer the user opportunities to explore items beyond their existing comfort zones. In such cases, the recommender should include items for which there is less existing evidence that the user will like them. This calls for the recommender to explore to what extent a user would be tolerant to exploration. In this thesis, we consider that different users will likely have different preferences with respect to item exploration. We propose personalized item filtering techniques for this, modeled under a multi-armed bandit framework, that consider (1) how and where exploration vs. exploitation items should be distributed in a result overview and (2) how adventurous exploration items can be, considering a user's general willingness to explore and existing familiarity with item categories in the collection. We present the results of a survey and an online quantitative experiment with 43 users of Muziekweb, a public music library that encourages taste broadening in the Netherlands, demonstrating the effectiveness of both our proposed filtering techniques that aim for personalized exploration.
Bachelor thesis
(2017)
-
Mitchell Dingjan, Rob Kapel, Maarten Sijm, Casper Karreman, Cynthia Liem, Ingmar Vroomen, Otto Visser
Muziekweb is the biggest source of music that has been released in the Netherlands. At this moment, all of their data is currently in Dutch. However, Muziekweb wishes their classical music titles to be translated to other languages as well. Since over 250.000 classical music titles exist, it is too much work to be done manually. Therefore, Muziekweb requested a program which can automatically translate titles. With these translations, not only can they deliver their data to more people, it will be easier in the future to link their data to other international sources.
Research has been done on these titles on how they are structured. This knowledge is required to split up these titles in smaller parts. Some of these parts can only have a limited set of Dutch words in them. Therefore, these translations can be translated using a lookup table. Besides these parts, some parts cannot be as easily translated, since they have a cultural history behind them. Therefore, these parts need to be translated differently. To do this, a connection is set up with external databases to find translations for these parts. To be able to improve this program in the future, the program is extensible as well, either for translating to more languages or linking up with more databases. ...
Research has been done on these titles on how they are structured. This knowledge is required to split up these titles in smaller parts. Some of these parts can only have a limited set of Dutch words in them. Therefore, these translations can be translated using a lookup table. Besides these parts, some parts cannot be as easily translated, since they have a cultural history behind them. Therefore, these parts need to be translated differently. To do this, a connection is set up with external databases to find translations for these parts. To be able to improve this program in the future, the program is extensible as well, either for translating to more languages or linking up with more databases. ...
Muziekweb is the biggest source of music that has been released in the Netherlands. At this moment, all of their data is currently in Dutch. However, Muziekweb wishes their classical music titles to be translated to other languages as well. Since over 250.000 classical music titles exist, it is too much work to be done manually. Therefore, Muziekweb requested a program which can automatically translate titles. With these translations, not only can they deliver their data to more people, it will be easier in the future to link their data to other international sources.
Research has been done on these titles on how they are structured. This knowledge is required to split up these titles in smaller parts. Some of these parts can only have a limited set of Dutch words in them. Therefore, these translations can be translated using a lookup table. Besides these parts, some parts cannot be as easily translated, since they have a cultural history behind them. Therefore, these parts need to be translated differently. To do this, a connection is set up with external databases to find translations for these parts. To be able to improve this program in the future, the program is extensible as well, either for translating to more languages or linking up with more databases.
Research has been done on these titles on how they are structured. This knowledge is required to split up these titles in smaller parts. Some of these parts can only have a limited set of Dutch words in them. Therefore, these translations can be translated using a lookup table. Besides these parts, some parts cannot be as easily translated, since they have a cultural history behind them. Therefore, these parts need to be translated differently. To do this, a connection is set up with external databases to find translations for these parts. To be able to improve this program in the future, the program is extensible as well, either for translating to more languages or linking up with more databases.