Machine-Learning-for-Personalization[1].pdf

Machine-Learning-for-Personalization

1. SUBMITTED BY : SASWAT MISHRA
REGD NO. : 2261020100
BRANCH : MCA
SEC : A
GUIDED BY : MR. BISWA MOHAN
ACHARYA

2. Various types of Personalization
Benefits of Personalization
Data Collection for content Personalization
MachineLearning Algorithms for Personalization
Real-World Examples of Machine Learning for Personalization
Challenges in Implementing Machine Learning for Personalization
Conclusion and Key Takeaways

3. •Content Personalization
•Product Recommendations
•Healthcare Personalization
•Service Personalization
•Contextual Personalization
•Adaptive Learning
Personalization comes in various forms, each catering to different aspects of user
experience and individual preferences. Here are several types of personalization

4. ?
Content personalization isa branding and marketing strategy in
which webpages, email and other forms of content are tailored
to match the characteristics, preferences or behaviors of
individual users.
For example, Netflix learns what users
like to suggest movies and shows based
on their interests.
So, if someone has watched a lot of
comedies in the past, it will probably
suggest more comedies for them. And
even provide an estimate of how likely
you are to enjoy those suggestions.

5. :
1Enhanced User Experience
Customizedrecommendations
andcontentimproveuser
satisfactionandengagement.
2Drives Business Results
3ImprovedCustomerLoyalty
Tailoredexperiencesfosterstrongerconnections
withcustomers,leadingtoincreased
loyaltyandadvocacy.
Researchshowsthat76%
ofconsumersaremore
likelytomakeapurchase
ifyourmessagingis
personalized.

6. Consider how compelling it is
for a customer to see products
on sale that are specifically
tailored to their preferences.
Like Amazon’s
recommendations

7. Serving them ads of the products
they viewed can jog their
memory and bring them back to
make a purchase.
Here’s an example of a
retargeting ad on Instagram
from Amazon

8. Consistently providing prospects and customers with
personalized content shows them they can trust your
brand. Making them more likely to become loyal
customers.
Why ???
Because customers prefer sticking with
a brand they trust rather than exploring
new (and potentially risky) options

9. :
Collaborative
Filtering
Recommend items
based on user
preferences and
similaritieswithother
users.
Content-Based
Filtering
Suggestitemsbasedon
theattributesand
featuresoftheitems
themselves.
Reinforcement
Learning
Optimize decisions
throughtrial-and-error
interactionswiththe
environment.

10. Recommender systems, also known as recommender engines, are
information filtering systems that provide individual recommendations
in real-time

11. Thecontent-based filtering(CBF) model provides recommendations using
specificattributes of itemsby findingsimilarities. Such systems create data
profiles relying on description information that may include characteristics of
items or users. Then the created profiles are used to recommend items similar to
those the user liked/bought/watched/listened to in the past.

12. The most commonly implemented model,collaborative filtering(CF) provides relevant
recommendations based oninteractionsof different users with target items. Such
recommender systems gather past user behavior information and then mine it to decide
which items to display to other active users with similar tastes

13. Collaborative
filtering
Memory basedModel based

14. •Thememory-basedCF method relies on the assumption that predictions can be
made on the pure “memory” of past data
•It uses the past ratings to predict the preferences of one user by searching for
similar users, or the “neighbors” that share the same preferences. That is why the
method is sometimes referred to asneighborhood-based.
•There are two ways to achieve memory-based
collaborative filtering
Item-item collaborative filtering
User-user collaborative filtering

15. makes predictions based on the similarities between items that users have
previously rated. Simply put, users get recommendations of items that are similar
to the ones they already rated taking into account ratings given by all users.
Amazon’s “Customers who bought items in your cart also bought”
recommendations are an example of item-item collaborative filtering.

16. •compares how different users rate the same items and in that way
calculates the similarity of their tastes.
•TikTok–the China-based social
media platform popular with
teenagers –recommends accounts
to follow with the help of user-
centered modeling.

17. •This type of CF usesmachine learning or data miningtechniques to
build a model to predict a user’s reaction to items
•The techniques used to build models include matrix factorization,
deep neural networks, and other types of ML algorithms.
matrix factorization
deep neural networks

18. The algorithm was a result ofthe 2009 Netflix Prize competition and
it is still one of the most popular filtering approaches in this field.

19. •It allows for modeling the
nonlinear interactions in the
data and finding hidden patterns
in the data that couldn’t
otherwise be discovered
•YouTube, for example, usesGoogle
Brain—an Artificial Intelligence
system —to drive relevant
recommendations of its videos.

20. •Spotifycombines
different
recommendation
models for
creating its
Discover Weekly
mixtape.

21. Data collection
explicit data implicit data
•Data storing
•Data analysis
•Data filtering —applying algorithms to make recommendations

22. Real-World Examples of Machine
Learning for Personalization
Netflix
Deliveringpersonalized
movieandTVshow
recommendationsbasedon
viewinghistoryanduser
preferences.
Spotify
Creating personalized
music playlistsand
recommendationstailored
toindividuallistening
habits.
Amazon
Offering personalized
productrecommendations
andtailoredcustomer
experiences.

23. :
1Data Qualityand Quantity
Adequateandrelevantdata
mustbecollectedand
processed for accurate
predictions.
2PrivacyandEthicalConcerns
Safeguardinguserprivacyand
ensuringethicaluseofpersonal
data.
3Algorithm Bias
Avoidingbiasandensuringfairnessinpersonalizedrecommendations.

24. :
Machinelearninghasthepowertotransformpersonalization,
deliveringtailoredexperiencesthatcaptivateandengage
users.Embracethepossibilitiesandunlockthefullpotentialof
personalizedexperiences.