This work explores the possibility of using relevant data from users’
social network to alleviate the cold-user problems in a recommender
system domain. The proposed solution extracts the most valuable
node in the graph generated by check in a venue with an Android
application using the Foursquare API. By obtaining the recommendations to this node we estimate the probability of some categories
to be similar to users tastes...
Alleviating cold-user start problem with users' social network data in recommendation systems
1. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Alleviating cold-user start problem with users’
social network data in recommendation systems
Eduardo Castillejo Aitor Almeida Diego L´pez-de-Ipi˜a
o n
DeustoTech - Deusto Institute of Technology, University of Deusto
http://www.morelab.deusto.es
Preference Learning: Problems and Applications in Artificial
Intelligence, 2012
2. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Index
1 Recommendation Systems
How do they work
What are they
2 Main problems of RS
Known problems
3 Proposed solution
Foursquare
Eigenvector centrality
Example and analysis
4 Results and evaluation
Evaluation
Results
5 Conclusions
Conclusions
Future work
6 Questions
3. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
How do they work
Amazon
4. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
How do they work
Youtube
5. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
What are they
Commonly built under a web-based platform they gather
information about every entity which takes part in an
e-commerce interaction process to make recommendations
to the users increasing the benefits of the e-commerce
company.
They use algorithms which base their recommendations in
explicit and implicit data from the users (ratings,
purchases, previous searches...).
6. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
Recommendation systems are a good tool to suggest items to
users based in their own interaction with the system, but they also
have some intrinsic problems which are difficult to solve:
Sparsity: it occurs when available data are insufficient for
identifying similar users (neighbors) and it is a major issue
that limits the quality of recommendations and the
applicability of collaborative filtering (CF).
Scalability: CF requires computations that are very
expensive and grow polynomially with the number of users
and items in a database. Therefore, in order to bring
recommendation algorithms effectively on the web, and
succeed in providing recommendations with high accuracy and
acceptable performance, sophisticated data structures and
advanced, scalable architectures are required.
7. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
Recommendation systems are a good tool to suggest items to
users based in their own interaction with the system, but they also
have some intrinsic problems which are difficult to solve:
Sparsity: it occurs when available data are insufficient for
identifying similar users (neighbors) and it is a major issue
that limits the quality of recommendations and the
applicability of collaborative filtering (CF).
Scalability: CF requires computations that are very
expensive and grow polynomially with the number of users
and items in a database. Therefore, in order to bring
recommendation algorithms effectively on the web, and
succeed in providing recommendations with high accuracy and
acceptable performance, sophisticated data structures and
advanced, scalable architectures are required.
Cold-start
8. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
The cold-start problem arises when a new entity enters the
system for the first time. In this situation the recommendation
engine can not predict suggestions because of the lack of
information about the current entity. It usually includes 3 entities:
Items
Users
Systems
9. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
Amazon
Recommendations related with Kindle, watches special prices
and laptops... ¿?
10. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
Youtube
Recommendations about videos of people we don’t actually
know... ¿?
11. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
We focus our research in alleviating the so called cold-user
problem by collecting information about the user digging in
their social network interactions.
12. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Known problems
We focus our research in alleviating the so called cold-user
problem by collecting information about the user digging in
their social network interactions.
But... how?
13. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Foursquare
Foursquare is a location-based social networking website
which allows users to ”check in” at venues using their
smartphones.
Thanks to its API developers can request some user data (e.g.
location, friends, last check-ins, etc.).
Developed Android app.
14. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Foursquare
Using the herenow API endpoint we get the previous
check-ins done by other users at the current checked in venue.
user 1 Legend
user 2
Friend user
Unknown user
1 hour interval
2 hours interval
3 hours interval
Users' check-ins time stamp
current user 1:00 pm
current user
user 1 2:00 pm
user 3
user 2 3:00 pm
user 3 1:00 pm
user 4 user 4 4:00 pm
15. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Eigenvector centrality
Once we have completed the matrix, we apply eigenvector
centrality.
Since degree centrality gives a simple count of the number of
ties a node has, eigenvector centrality acknowledges that
not all connections are equal.
Denoting the centrality of a node i by xi , then it is possible
to make xi proportional to the average of the centralities of i’s
network neighbours:
where λ is a constant. This equation can be also rewritten
defining the vector of centralities x = (x1 , x2 , ...):
16. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Example and analysis
Applying eigenvector centrality to our previous matrix A”
we obtain that the eigenvalue λ = 12,502, and its eigenvector:
17. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Example and analysis
The first value of the vector e1 is related to the node 0 (the
current user), so its value has to be ignored (in this case the
highest value corresponds with the current user).
The second highest value corresponds with the node 1. That
means that his recommendations would be more ”pleasing” to
the user.
18. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Evaluation
Amazon default recommendations VS. our Amazon
categories estimation: We presented to our test users
the default recommendations that Amazon.com for new
users, and another list with our Amazon categories
recommendations computed with our solution.
Once our users compared both lists, they fulfilled a
questionnaire to capture their satisfaction level with the
obtained results.
Amazon default recommendations: Kindle related products,
clothing trends, products being seen by other customers, best
watches prices, laptops best prices, top seller books.
Amazon default categories: Home, garden and tools,
clothing, shoes and jewelry; books; electronics and computers;
automotive and industrials; movies, music and games; grocery,
health and beauty; toys, kids and baby; sports and outdoors.
19. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Results
20. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Conclusions
This paper explores the possibility of using relevant data from
users’ social network to alleviate the cold-user problems in a
recommender system domain. The proposed solution extracts
the most valuable node in the graph generated by check in a
venue with an Android application using the Foursquare API.
By obtaining the recommendations to this node we estimate
the probability of some categories to be similar to users
tastes...
21. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Conclusions
This paper explores the possibility of using relevant data from
users’ social network to alleviate the cold-user problems in a
recommender system domain. The proposed solution extracts
the most valuable node in the graph generated by check in a
venue with an Android application using the Foursquare API.
By obtaining the recommendations to this node we estimate
the probability of some categories to be similar to users
tastes...
... but we suffered several limitations:
Few users and data...
Near venues are not checked in enough...
22. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Conclusions
This paper explores the possibility of using relevant data from
users’ social network to alleviate the cold-user problems in a
recommender system domain. The proposed solution extracts
the most valuable node in the graph generated by check in a
venue with an Android application using the Foursquare API.
By obtaining the recommendations to this node we estimate
the probability of some categories to be similar to users
tastes...
... but we suffered several limitations:
Few users and data...
Near venues are not checked in enough...
Sparsity.
23. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Future work
Include data not only from Foursquare
Combine different social network analysis metrics
Take into account more than the most valuable node for
doing recommendations.
Store the obtained matrices for each venue and update them
with every check-in.
Test the solution among a higher number of users.
24. Recommendation Systems Main problems of RS Proposed solution Results and evaluation Conclusions Questions
Thank you again!
eduardo.castillejo@deusto.es
Preference Learning: Problems and Applications in
Artificial Intelligence, 2012