Semantic-based Reconstruction of User’s Interests in Distributed Systems
Abstract: Generally, the user requires customized data reflecting his current needs represented in terms of interests that are stored in his profile. Therefore, taking into account user’s profile is significant to improve the returned results. Day by day, the user becomes more and more active in social networks and uses different distributed systems. In this context, the problem is that the access to user’s interests becomes more and more difficult mainly after updating and/or enriching the user’s profile. This may produce cognitive overload problem, which is time consuming in terms of browsing the user’s profile. This problem can be solved by reorganizing user’s interests. Most of the proposed reorganization methods use machine learning algorithms and different similarity measures. As the user’s interests are characterized by their popularity and freshness, other approaches combine these characteristics into the notion of temperature in order to keep in the profile uniquely the corresponding interests for a period of time. In this paper, we propose an approach to reconstruct the user’s profile by taking into account the semantic relationships between interests and by respectively merging the temperature and the k-means learning algorithm.
| Main Authors: | , , , , , |
|---|---|
| Format: | Digital revista |
| Language: | English |
| Published: |
Instituto Politécnico Nacional, Centro de Investigación en Computación
2017
|
| Online Access: | http://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1405-55462017000300545 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract: Generally, the user requires customized data reflecting his current needs represented in terms of interests that are stored in his profile. Therefore, taking into account user’s profile is significant to improve the returned results. Day by day, the user becomes more and more active in social networks and uses different distributed systems. In this context, the problem is that the access to user’s interests becomes more and more difficult mainly after updating and/or enriching the user’s profile. This may produce cognitive overload problem, which is time consuming in terms of browsing the user’s profile. This problem can be solved by reorganizing user’s interests. Most of the proposed reorganization methods use machine learning algorithms and different similarity measures. As the user’s interests are characterized by their popularity and freshness, other approaches combine these characteristics into the notion of temperature in order to keep in the profile uniquely the corresponding interests for a period of time. In this paper, we propose an approach to reconstruct the user’s profile by taking into account the semantic relationships between interests and by respectively merging the temperature and the k-means learning algorithm. |
|---|