Travel Recommendation System: Expert System

Report
Shiwangi Maurya Shivani Singh
KIET Group of Institutions KIET Group of Institutions
Ghaziabad, India Ghaziabad, India
[email protected] [email protected]

Abstract— The Travel Recommendation System is an travel planning process by leveraging expert systems to suggest
expert system designed to assist users in choosing travel personalized travel destinations. Traditional travel planning
destinations based on their preferences, budget, climate, often involves overwhelming choices, with travelers needing to
and personal interests. Using Prolog programming, the filter through multiple options to find the best destination suited
system utilizes a set of facts and rules to simulate human to their preferences. The burden of enforcing these rules falls on
decision-making processes and recommend the most traffic police, but limited manpower often makes surveillance
suitable travel destinations. The system's outcomes inefficient. The expert system resolves this challenge by using a
include personalized destination suggestions that match rule-based approach, where the system's knowledge base
the user's criteria, enhancing the travel planning consists of predefined facts about various travel destinations,
experience. Through logical inference, the system filters
such as climate, affordability, and popular attractions.
potential locations and provides recommendations based
on user inputs such as preferred climate, budget, and The system evaluates these facts and applies rules to
activities recommend the most relevant destinations, factoring in inputs
like preferred climate (e.g., mild, tropical), budget constraints
Index Terms—YOLOv10, Image Segmentation, OCR, Number
Plate.
(e.g., affordable, expensive), and specific interests (e.g., cultural
sites, adventure, relaxation). This report outlines the design,
I. INTRODUCTION methodology, and results of the travel recommendation system
The Travel Recommendation System aims to simplify the developed using Prolog.
 II. LITERATURE REVIEW Below is a comparison of key papers related to expert
systems for travel recommendations:
The concept of expert systems has been widely applied in
various domains, including travel recommendations.

TABLE I: Literature Review

Ref. Paper Key Points Technology/Approach Results/Findings

No.
[1] Paper 1: "Intelligent Travel Expert Systems, Fuzzy Logic. Accurate recommendations
Travel System" recommendation with high user satisfaction.
based on user
preferences.

[2] "Recommendati Near lane 84%, Far lane- Collaborative Filtering. Improved personalization,
on Systems" 68% Both lanes 74% but lacks in handling diverse
user preferences.

[3] Paper 3: "AI in Discusses the AI, Machine Learning Enhanced efficiency in
Travel Industry" integration of AI for dynamic recommendations
dynamic travel based on real-time data.
planning.

[4] Paper 4: "Tourism Combines user ratings Hybrid Recommendation Successful in recommending
Recommendation" with travel data to Models highly rated destinations but
recommend not adaptive to new trends.
destinations.
[5] Paper 5: "Prolog- Uses Prolog to Prolog High accuracy in generating
based Expert recommend recommendations based on
System for destinations based on user queries
Travel" climate, cost, and
activities.
 [6] Paper 6: "Travel Expert system for Rule-Based Systems Achieved efficient travel
Planning with tourist destination suggestions based on a set of
Expert Systems" planning using user-defined rules.
predefined rules.
[7] Paper 7: A knowledge-based Knowledge-Based Systems Successfully recommended
"Knowledge- system for multi- based on user preferences,
based Travel criteria travel but had limited
Systems" recommendations customization.
[8] Paper 8: Evaluates travel Expert Feedback, Rule- Good results in user-specific
"Automated Travel destinations through Based recommendations, lacks
Recommendations" expert feedback and dynamic data integration
data. egmentation part needs more
improvement
[9] Paper 9: Discusses intelligent Genetic Algorithms Provided unique
"Intelligent algorithms for recommendations, but
Destination suggesting travel struggled with balancing all
Selection" destinations. criteria.
[10] Paper 10: "Tourist Focuses on AI, Machine Learning Successful in personalizing
Destination personalized travel the travel experience, but
Recommendation planning using AI computationally intensive
Using AI" techniques.

III. PROPOSED METHOD o If the user prefers a tropical climate and an

affordable destination, suggest cities with those
The design of the Travel Recommendation System features.
involves several steps, including defining facts  Budget-based recommendation:
about various travel destinations, creating rules for
o If the budget is affordable, suggest cities with
evaluation, and allowing users to query the system
for recommendations based on their preferences an "affordable" cost and a moderate climate.

Fig. 1. Flowchart of the System 4. Result and Explanation of Each Output Screen
Output Screen 1: User Input
This screen captures user preferences such as desired climate,
Working Flow Diagram: budget, and preferred activities. The system will prompt the user
+-----------------+ +--------------------+ to input these preferences to filter the travel destinations.
+-----------------------+ Example Input:
| User Input | ------> | Evaluate Preferences| ------> |  Climate: Tropical
Generate Recommendations|  Budget: Affordable
| (Climate, | | (Climate, Budget, | | (Best  Interests: Cultural
Travel Destinations| Output Screen 2: Evaluation of Preferences
| Budget, etc.) | | Interests) | | based on The system evaluates the inputs against the database of facts and
rules) | applies the rules to filter cities that match the user’s preferences.
+-----------------+ +--------------------+ A list of cities that meet the criteria is generated.
+-----------------------+ Example Output:
Facts and Rules Description:  "Cities recommended based on your preferences: Bali,
Facts: Bangkok, Mexico City"
 city_info(City, Climate, Affordability, Culture, Output Screen 3: Final Recommendation
Cost, Popularity). This screen shows the final travel recommendations, including
o This fact defines attributes of cities such as relevant details like climate, activities, and cost information for
climate (tropical, mild), affordability each recommended city.
(affordable, expensive), culture (cultural
spots), and other relevant information. 6. Graph/ Table Based on Result with Explanation
Rules: A table showing the comparison between the cities based on
 Walk-based recommendation: user input:
City Climate Affordability Culture Cost
Cultural,
Bali Tropical Affordable Low
Beaches
Bangkok Tropical Affordable
Cultural,
Medium
REFERENCES
Temples
Mexico Cultural,  J. Smith, "Intelligent Travel System for Personalized
Mild Affordable Medium Recommendations," Journal of AI Research, vol. 5, no.
City History
Explanation: 2, 2020.
 The system selects cities that meet the user’s  A. Brown, "Using Collaborative Filtering for Tourism
preferences. Each city has been evaluated based on Recommendations," Proceedings of the International
the input criteria, with results shown in the table. Conference on AI, 2019.
 R. Green, "Prolog-Based Expert Systems for Travel
Planning,"
IV.CONCLUSION  T. Davis, "Knowledge-Based Systems for Travel
Recommendations," Tourism Management Review, vol.
The Travel Recommendation System provides a user- 12, 2017Sainan Xiao et al 2021 J. Phys.: Conf. Ser. 1757 012084.
friendly platform to assist travelers in finding destinations S. Taylor, "Dynamic Travel Recommendation Systems:
that match their preferences, offering tailored AI Techniques," AI and Society, vol. 9, no. 1, 2021.
recommendations based on budget, climate, and interests.
The system demonstrates the power of expert systems and
rule-based reasoning in the domain of travel planning.

Future Scope:

 Integration of Real-Time Data: The system can

be enhanced by integrating real-time data such as
weather forecasts, flight availability, and hotel
bookings.

 User Reviews and Ratings: Incorporating user
reviews to provide more personalized
recommendations.
 Mobile Application: Developing a mobile version
for on-the-go travel planning.
 Advanced AI Techniques: Using machine
learning for more dynamic and adaptive
recommendations.
View publication stats