ICACT20240453 Slide.16
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Than You
|
|
ICACT20240453 Slide.15
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Conclusion
Recommendation systems are a way to stay ahead of the curve and achieve larger business goals like increasing sales, ad revenue, and user engagement.
Recommendation systems must adapt to remain relevant and effective as user behaviors, preferences, and needs change.
Agility is also essential for recommendation systems.
Agile recommendation systems evolve, considering what works and what doesn't and additional data sources that help improve recommendations.
By continually evaluating and enhancing AI- powered recommendation systems, businesses like health-Care can achieve their goals while providing their users with the best possible experience.
|
|
ICACT20240453 Slide.14
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Challenges in Recommender Systems
Cold start problem
Synonym
Privacy
Sparsity
Latency problem
Shilling attacks
Grey sheep
Evaluation and the availability of online datasets
Context-awareness
|
|
ICACT20240453 Slide.13
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Case study: online reservation Health-Care phases
|
|
ICACT20240453 Slide.12
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Case study flowchart: online reservation Health-Care
|
|
ICACT20240453 Slide.11
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Benefits of using AI-powered recommender systems(2)
Minimizing frustration in the customer experience
By filtering and displaying personalized and relevant content to customers
A recommendation engine can analyze consumer behavior and preferences to make recommendations that engage and intrigue customers, improving the customer experience and increasing sales and income for organizations.
Meet customer experience expectations:
The recommendations are tailored specifically for the individual based on their past behavior and preferences.
By analyzing customer behavior and preferences, these engines can provide tailored product recommendations anticipating their needs and desires
Boosting Business performance with recommendation engine:
By showing personalized product recommendations to the customers.
With the help of an AI solutions provider, you can efficiently integrate a recommendation engine into your business operations and enjoy the benefits of increased sales, higher average order value, and better revenue growth.
|
|
ICACT20240453 Slide.10
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Benefits of using AI-powered recommender systems(1)
Personalization
Can help by providing customers with relevant, tailored recommendations that meet their unique need and preferences.
The recommendations made by the AI-based recommendation system can help customers discover new products they may be interested in. One way to save time is by choosing items that one is more likely to purchase.
Unified customer experience across channels:
AI-powered recommendation system can be a game-changer to provide a consistent experience across all channels, including social media, websites and mobile apps.
Users receive a similar experience everywhere by collecting and utilizing data from customer interaction on different channels
Deliver Relevant Content:
Delivering relevant content has become key to customer engagement.
This approach ensures consumers only see information they like, increasing engagement and satisfaction.
By evaluating search searches, purchases, and browsing history AI-powered recommendations can suggest stuff users are more likely to appreciate.
|
|
ICACT20240453 Slide.08
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Figure
|
|
ICACT20240453 Slide.07
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Figure
|
|
ICACT20240453 Slide.06
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Phase of the Recommender Systems.
Information collection:
Gather user data and generate a profile or model with metadata, behavior, and attributes..
Recommender systems use explicit feedback, implicit feedback, or hybrid feedback to determine user preferences.
Learning phase:
A learning algorithm to filter
Using user attributes from prior responses.
Prediction/ Recommendation:
Predict/recommend user-liked things.
memory-based or model-based techniques applied on the collected data Can be achieved based on the previous activities of the user.
|
|
ICACT20240453 Slide.05
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
applications of recommender systems in healthcare
The emerging healthcare websites and online medical databases impose a new challenge for medical service recommendation
Model-based collaborating filtering is a popular technique used in healthcare domain
It is uses to better understand the interactions among different genes within the biological system
Matrix factorization like iDoctor is the professionalized doctor recommendation system.
hybrid matrix factorization is used to propose a topic model-based strategy to identify user preference distribution and doctor feature distribution for more accurate and tailored medical suggestions.
An example of a content-based filtering in healthcare that link healthcare consumers to reputable healthcare educational websites form MedlinePlus for a given health video from YouTube
|
|
ICACT20240453 Slide.04
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Different types of recommender systems
Content-based uses the content information of the items for building the recommendation model .
Collaborative filtering : is a technique used to make automatic predictions about the preference of a user derived by analyzing the preferences of other users that have similar interests
Hybrid methods: this recommendation is built by integrating many recommender systems for greater reliability
For example, Collaborative filtering, which fails when new things don't have ratings, can be used with content-based systems, which provide feature information.
|
|
ICACT20240453 Slide.03
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Different types of recommender systems
To bring out valuable insight from the massive amount of data with speed and precision, recommender systems rely on various data mining techniques for information retrieval
The one criterion that is used to classify them is the filtering technique : Collaborative, Content-based and hybrid
Collaborative filtering is based on trust and emotion. It can be slit into many categories : model-based, memory based, clustering based, matrix and tensor factorization etc..
|
|
ICACT20240453 Slide.02
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Recommender Systems
Definition:
Recommendation systems play a pivotal role in enhancing user experiences by providing personalized suggestions tailored to individual preferences.
These systems have become an integral part of various online platforms, from e-commerce and streaming services to social media, news, and Healthcare recommendations.
The power of recommendation systems lies in their ability to:
To anticipate user needs.
Leading to increased engagement.
Customer satisfaction.
Business growth.
|
|
ICACT20240453 Slide.01
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Index
Recommender Systems
Different types of recommender systems
The phase of the Recommender Systems
Benefits of using AI-powered recommender systems
Challenge &Conclusion
|
|
ICACT20240453 Slide.00
[Big slide for presentation]
|
Chrome Text-to-Speach Click!! |
 |
Enhanced Experiences: Benefits of AI-powered Recommendation Systems
Department of Digital Anti-aging Healthcare,
INJE UNIVERSITY
Kouayep Sonia Carole, Tagne Poupi Theodore Armand and Hee-Cheol Kim
|
