2. Intended Learning OutcomesIntended Learning Outcomes
of this courseof this course
Intended Learning OutcomesIntended Learning Outcomes
of this courseof this course
3. Researchers’ troubles
•Every day, a researcher:
– Spend about half of their
working time just searching
for information
– Do not receive timely update
information related to their
research
– Find uneasy to connect with
other researchers for joint
research activities
4. Everything is changing
• Based on above problems of researchers ,
recommendation techniques ‘s coming will
have great influence in all aspects of our life.
Traditional future
Recommendation
techniques
5. What’s the recommendation
techniques ?
Recommender techniques
are information agents that
attempt to predict which
items out of a large pool a
user may be interested in
and recommend the best
ones to the target user.
6. Techniques category
• The techniques can be classified based on the information
sources they use .
• The available sources are the user features (demographics)
(e.g. age, gender, income, location), the item features (e.g.
keywords, genres), the user-item ratings (explicit ratings,
transaction data) and knowledge about user and item(for
reasoning).
10. Content-based recommendation
Content-based recommendation
methods use the information about
item features and the ratings a
user has given to items.
The technique combines these ratings
to a profile of the user’s interests
based
on the features of the rated items.
Maybe the clerk advises
you to buy some
trousers according to
your styles and
preferences
11. Collaborative filtering recommendation
The users are
categorized based on the
attributes of their demographic
profiles or on similar rating
preferences in order to find users
with similar
features.
The technique then recommends
items that are preferred by these
similar users
Maybe you will receive
suggestions from you
like-minded friends
(similar demographic
profiles or similar
preference)
12. Knowledge-based recommendation
Maybe you will take the
recommendations considering
the knowledge about price
,quality and so on.
Considering the users’
specific tasks,
Knowledge-based
recommendation can
address this problem by
using a model of
knowledge.
15. Thank you for your
attention
If you want to know more
about this interesting
techniques , please wait
for next class!
Notes de l'éditeur
These items can be of any type, like movies, music, books,
Websites , commodities or collaborators. The user’s interest in an item is expressed through the rating the user gives
the item. A recommendation system has to predict the ratings for items that the user has not yet seen.