Recommender Systems

.
RECOMMENDER SYSTEM
WHAT ARE RECOMMENDER
SYSTEMS?
INTRODUCTION:
•An application of AI
•Technology that predicts user preferences and suggests relevant
items.
•Recommender systems are algorithms providing personalized
suggestions for items that are most relevant to each user.
•Used in various domains: e-commerce, music streaming, video
platforms , Social media.
•Enhances user experience and drives engagement.
Recommender Systems (RS)
E-commerce
Application areas:
• E-commerce. e.g Amazon,Daraz.
• Video platforms. e.g Netflix,Youtube.
• Music Streaming. e.g
Spotify,Soundcloud.
• Social Media. E.g
Instagram,Facbook,Twitter
Recommender Systems (RS)
Music streaming(Spotify)
Social media(Instagram)
Video streaming(Netflix)
Why Recommender Systems?
Value for the customer
• Find things that are interesting
• Narrow down the set of choices
• Help me explore the space of options
• Discover new things
Value for the provider

• Additional and probably unique personalized service for the customer
• Increase trust and customer loyalty
• Increase sales, conversion etc.
• Opportunities for promotion, persuasion
• Obtain more knowledge about customers
Why Recommender Systems?
Youtube: Every minute people upload 500 hours Tentative research:
• Amazon.com generates 30 to 70%
of videos, i.e. it would take 82 years to a user to percent of their sales through the
recommendation lists
watch all videos uploaded just in the last hour.
• Netflix (DVD rental and movie
Spotify: Users can listen to more than 80 million streaming) generates 50 to 70%
percent of their sales through the
song tracks and podcasts. recommendation lists
Amazon: Users can buy more than 350 million

different products.
How RS Works?
Input:
• User : History,rating,preferences,situational context
• Items/content (with or without description of item
characteristics)
Output:
• Recommend items that are assumed to be relevant
CLASSIFICATION OF RS
Recommender system can be classified according to the kind of information used
to predict user preferences as Content-Based , Collaborative Filtering and

Hybrid Method.
8
Collaborative Filtering (CF)
Collaborative filtering methods use the feedbacks or activity history of all users in order to
predict the rating of a user on a given item by inferring interdependencies between users
and items from the observed activities.
• Basic assumption and idea

– Users give ratings to catalog items (implicitly or explicitly)
– Customers who had similar tastes in the past, will have similar tastes in the future
Collaborative Filtering (CF)
• Explicit Feedback:
– Explicit feedback is a rating explicitly given by the user to express their satisfaction
with an item. Examples are: number of stars on a scale from 1 to 5 given after buying a
product, thumb up/down given after watching a video, etc. This feedback provides
detailed information on how much a user liked an item, but it is hard to collect as most
users typically don’t write reviews or give explicit ratings for each item they purchase.
• Implicit Feedback:
– Implicit feedback, on the other hand, assume that user-item interactions are an
indication of preferences. Examples are: purchases/browsing history of a user, list of
songs played by a user, etc. This feedback is extremely abundant, but at the same time it
is less detailed and more noisy (e.g. someone may buy a product as a present for
someone else). However, this noise becomes negligible when compared to the sheer size
of available data of this kind, and most modern Recommender Systems tend to rely on
implicit feedback.
CONTENT BASED APPROACHES:
• Content Based Methods:

– Content-based methods describe users and items by their known metadata.
– This METADATA typically describes various attributes or features of the items,
which the system can use to analyze and compare them to the user's preferences.
– For example, in a movie recommendation system, metadata could include attributes
such as genre, director, actors, release year, plot keywords, language, and ratings.
Similarly, in a music recommendation system, metadata might include attributes like
artist, album, genre, year of release, duration, and musical features such as tempo or
mood.
– By analyzing the metadata associated with items, the content-based recommender

system can generate recommendations that match the user's preferences based on
similarities between the metadata of items they have liked or interacted with and the
metadata of other items in the system
HYBRID METHOD:
• Hybrid Method:
– Finally, there are also hybrid methods that try to use both the known metadata and the
set of observed user-item interactions. This approach combines advantages of both
Content-Based and Collaborative Filtering methods, and allow to obtain the best
results.
HYBRID METHOD:
Hybrid: combinations of various inputs
and/or composition of different mechanism
CONCLUSION:
Recommender systems have revolutionized the way we interact with online
platforms. By personalizing the user experience, they help us discover
relevant content, products, and services. As artificial intelligence and data
science continue to evolve, recommender systems will become even more
sophisticated, offering hyper-personalized recommendations. However, it's
important to address ethical considerations surrounding user privacy,
transparency in recommendation algorithms, and mitigating potential biases
within the data. By ensuring responsible development and deployment,
recommender systems can continue to be a powerful tool for fostering user
engagement and satisfaction in the digital age.

Recommender Systems

Uploaded by

Recommender Systems

Uploaded by

.

Value for the provider

Amazon: Users can buy more than 350 million

to predict user preferences as Content-Based , Collaborative Filtering and

• Basic assumption and idea

• Content Based Methods:

– By analyzing the metadata associated with items, the content-based recommender

You might also like