2. Which places should I Visit
I would like to visit a place with nature. Places where I feel
like I am in heaven.
I want to be away from the work and experience a lot of
adventure, trekking and hiking.
Let’s go to explore the mythical places. History and myths
are what fascinates us.
4. Beginning
• A recommender system
• Suggests different places to the tourists
• Uses the characteristics and history of the users/tourists
5. Purpose/Objectives
• Self Agent
• Information before visit
• Explore hidden treasures/places of Nepal
• Promote tourism
6. Challenges
• List the intended outcomes for this training session.
• Each objective should be concise, should contain a verb, and
should have a measurable result.
• Tip: Click and scroll in the notes pane below to see examples,
or to add your own speaker notes.
11. Collaborative Filtering
• Widely-used recommendation approach
• Prediction the utility of items for a user
▫ Matrix Factorization Model
▫ Association rules
▫ Nearest-Neighbor
12. Nearest-Neighbor
• Memory-based approach
• Utilizes the entire user-item
• Approach includes
▫ User-based methods
▫ Item-based methods
13. Association rules
• Each transaction for association rule mining is the set of
items bought by a particular user.
•We can find item association rules, e.g.,
visit_X, visit_Y -> visit_Z
14. Matrix Factorization Model
• Map both users and items
ȓ푢푖= 푞푖
푇푝푢 (1)
• 푞푖 & 푝푢 are the vectors of item and users ȓ푢푖 is rating item ‘i‘
• Factor vectors (푝푢 and 푞푖 ), minimizes the regularized squared
error on the set of known ratings:
푚푖푛
푞∗, 푝∗
(푢,푖)∈푘
(푟푢푖 − 푞푖
푇푝푢)2+휆(||푞푖 ||2 + ||푝푢||2) (2)
15. Implementation
• Data Collection
• Implementing Recommendation
• Stakeholder Analysis
• Market For Recommendation System
16. Data Collection
• Source:
▫ Nepal Tourism Board
▫ Ministry of Tourism, Culture and Civil Aviation
17. Implementing Recommendation
• New Users
▫ Those who haven’t visited any place in Nepal
▫ Based on their characteristics: Nationality, Age Group &
Gender
• Existing Users
▫ Those who have already visited some places in Nepal
▫ Based on their history of visiting places
18. Stakeholder Analysis
• Identifying all the stakeholders
• Prioritizing Stakeholders
• Understanding Stakeholders
• Stakeholders Involvement
19. Market For Recommendation System
• Establish/Maintain the communication with the
customers/users
• Business Model
20. Limitations
• Time to process recommendation is comparatively high
• Focused only on foreign tourists
• Lack of complete information about the places
21. Future Enhancements
• Make the service for Nepalese
• Faster data processing
• Complete information about every tourist place in
Nepal
• Tourist service recommendation
• Path to the destination
22. Future Enhancements contd..
• User generated content and social networking services
•Multiple days tour planning
• Intelligent UI
How presentation will benefit audience: Adult learners are more interested in a subject if they know how or why it is important to them.
Presenter’s level of expertise in the subject: Briefly state your credentials in this area, or explain why participants should listen to you.
Self Agent:- Help the tourist to get the information about the place by themselves
Information before visit:- Tourist can find the way to plan the trip
Promote Tourism:- Help to promote the tourism industry in Nepal
Example objectives
At the end of this lesson, you will be able to:
Save files to the team Web server.
Move files to different locations on the team Web server.
Share files on the team Web server.
JAVA for class building to make it object oriented system
JSP for web page presentation
PHP for visualization
MySQL for database processing
MySQL Workbench for database designing
JavaScript for client side verification purpose
CSS for designing
D3 for graphical presentation
predicts the utility of items for a user based on the items previously rated by other like-minded users.
predicts the utility of items for a user based on the items previously rated by other like-minded users.
utilizes the entire user-item database to generate predictions directly, i.e., there is no model building.