Artificial Intelligence based Personalized Travel itinerary planner: A Review by IRJET Journal

Artificial Intelligence based Personalized Travel itinerary planner: A Review

Priya Raj1 , Prabha Suman2 , Mahesh G3 , Sur Singh Rawat4, Gyanendra Kumar5

1Student, Dept. of CS Engineering, JSS Academy of Technical Education Noida, Uttar Pradesh, India

2Student, Dept. of CS Engineering, JSS Academy of Technical Education Noida, Uttar Pradesh, India

3Professor, Dept. of CS Engineering, JSS Academy of Technical Education Noida, Uttar Pradesh, India

4Professor, Dept. of CS Engineering, JSS Academy of Technical Education Noida, Uttar Pradesh, India

5Professort, Dept. of IOT and Intelligent System, Manipal University Jaipur, Rajasthan, India

Abstract - The travel itinerary maker provides offline accessibility, expense management capabilities, andflexibility to adjust to unexpected scenarios. The tool uses artificial intelligence to provide personalized recommendations, streamline route planning, andenhance the travelexperience. Future integrations, like wearables and AR/VR technology, promise to enhance the user experience further.

This project promotes sustainable travel, responsible exploration, and intelligent trip planning. The travel itinerary maker allows customers to have easy, efficient, and enjoyable travel experiences.

Key Words: smart travel planner, artificial intelligence, personalization, dynamic scheduling, navigation

1.INTRODUCTION

Travel planning is a complex process that includes researching places, comparing itineraries, and adapting arrangementstopersonaltastes.Artificialintelligenceoffers personalizedandeffectivewaystoaddressthesedifficulties. Thistravelitinerarygeneratorcreatestailoredandefficient plans for users based on their destinations, preferences, length,andcurrentconditions.Wedevelopthreeevaluation criteria for the planner-generated trip itineraries: Plausibility,Completion,andPersonalization[4]

 Rationality -Learnhowtomodelconstraintsintravel scenariosandbuildsensibleroutesaccordingly.

 Completeness -Howtooffercomprehensivetravelservices,includingguidanceandplanning,foraccurateand entertainingitineraries.

 Personalization -Howtoidentifyandexploitimplicit information about user personalization to deliver individualizedrecommendationsandserviceplanning.

1.1 Motivation of the Study

Traveling is a pleasant experience, but preparation can be time-consuming and difficult, especially for those with restrictedbudgetsanddifferentinterests.Challengesinclude:

1) Information overload: The number of vacation alternatives, recommendations, and reviews might makechoosingtough.

2) Customizable distances: Generic packages do not accommodateindividualtastes.

3) Dynamic constraints:Budgetlimits,pricevariations, andunexpectedchangesincreasethedifficulty.

4) Time-consuming: Researching, researching, and coordinatingtripplansrequiressignificanteffort.

5) Connectivity problem solved:Offlinefeatureenables easy travel in locations with limited internet connectivity.

Thisprojectaimstosimplifytripplanningandmakeitmore accessible,allowingpeopletoenjoythejourneyratherthan the preparation process. We hope to revolutionize how individuals discover, organize, and enjoy their travel experiencesbyincorporatingcutting-edgeAIapproaches

1.2 Organization of Study

The paper is structured as follows: The introduction discussestheimpetusfordevelopinganAI-basedpersonalized trip itinerary planner, the obstacles of traditional travel planning,andthestudy’saims.

The literature study gives an overview of current travel planningsystems,emphasizingtheirshortcomingsandthe potentialforAItechnologytoaddresstheseissues.

TheMethodologyandSystemDesignsectiondiscussesthe proposedsystem’sarchitecture,includingtheintegrationof hardware, software, and data sources like travel APIs, weather data, and user profiles. The Personalised and Algorithmic Approach section delves into the machine learningmodelsandoptimizationapproachesusedtoadjust routes to specific user preferences, as well as the system workflow.

ThePerformanceMonitoringandEvaluationsectioncovers thecriteria.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

Table -1: Listofsymbolsandabbreviations

Abbreviation Description

AI ArtificialIntelligence

AR AugmentedReality

VR VirtualReality

API ApplicationProgrammingInterface

ML MachineLearning

IoT InternetofThings

NGSAII Non-dominated Sorting Genetic AlgorithmII

PCA PrincipalComponentAnalysis

HTTP HyperTextTransferProtocol

RNN RecurrentNeuralNetwork

MQTT Message Queue Telemetry Transport Protocol

LSTM LongShort-TermMemory

FedAvg FederatedAveraging

MCTS MonteCarloTreeSearch

MARL Multi-AgentReinforcementLearning

GA GeneticAlgorithm

2. LITERATURE REVIEW

A summary of relevant literature in English An in-depth reviewofexistingresearchhasrevealedmajorinsightsand breakthroughs in the field of travel itinerary planning, particularly through the use of artificial intelligence (AI) approaches.Toaddresstheseproblems,thisliteraturereview willpro-videadetailedoverviewofcurrentresearchonAIbasedtravelitineraryplanning.Itwilllookatthemostrecent developmentsinpersonalizedrecommendersystems,realtime data integration approaches, strong data privacy frameworks, and efficient algorithms for optimization for travelitineraryplanning.Furthermore,itwilllookintohow emergingtechnologiesliketheInternetofThings(IoT)and artificial intelligence [6] [8] might be used to increase the precisionandadaptabilityoftravelitinerarysolutions

2.1 State-of-the-art

• AutomatePersonalized ItineraryCreation:Users’ se-lectionswillbuildatripitineraryfortheirchosen destination, saving them time and providing a scheduletolookforwardto[1].

– Energy-efficient routing maps: These campaignstrytosaveenergybydeter-mining themostefficienttravelroutes.[2]

– AI-basedchatbotservices:Itusesthetou-rism dataNER(NamedEntityRecognition)andDST (Dialogue State Tracking) models, which perform transfer learning of the alreadytrained language models (PLMs), and the tourisminformationknowledgebaseofNeo4J graphdatabase.[1]

• Enhance Travel Flexibility and Navigation: Itwill allowcustomerstocreatemoreflexibleandscalable tourschedules.[8][1][2]

– Dynamic Scheduling: The platform may automaticallyadaptthescheduledepen-denton thesizeofthetourcompany.

– Solo-Friendly Itineraries: For those trav-eling alone,theplannercanoffersolotourprograms thatfocusonreviewstailoredtosingletourists, such as self- guided tours and personalized culturaltours.

– Dynamic Transport Solutions: Solo or group travelerscanreceiveoptimizedtourroutesand adviceforspecificshippingmodes.

• Scalability:Byfocusingonthesescalabilityfactors, a personalizedrouteitineraryplannercanquickly expand while maintaining an excellent user experience.

– Monitoring and Analytics: Scalable journey architecture should include reliable tracking equipmenttomonitortheoveralloutcomeand consumerbehavior.Scalableanalyticssystems (suchasGoogleAnalyticsandMixpanel)should beincludedtodeliverreal-timeinsightsinto usage patterns and assist in enhancing the consumerexperience.

– Global Expansion: To assist consumers internationally, a personalized route planner should be created to handle a variety of languages,currencies,andlocalneeds.

– Third-Party Integrations: As more destinations, activities, and products are introduced,theplannerwillwishtointegrate with various third-party APIs (hotels, flight booking,nearbyevents,andtransportation).

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

• Real-time Performance:Real-timeoverallperformancereferstotheabilityofatripitineraryplanning gadget to dynamically react to changing situations,providingconsumerswithtimelyupdates andoptimizeditineraries.Thisfeatureensuresthat travelers may effectively negotiate unexpected conditions, boosting comfort and overall pleasure [7][14]. –

Traffic and Transportation Updates: Traffic congestion and transit delays all have an impactonthefeasibilityofarrivingontime. –

Real-TimeBookingandAvailability:Ensuring theavailabilityoflodgings,transportation,and sportsiscrucialforasuccessfultour.

• Environmental Factors: Environmental factors have a significant impact on tour itineraries, impactingtheplanning,execution,andenjoymentof thetrip[11].

– DynamicAdaption:Anotherimportantaspect isthatitineraryplansareupdatedinrealtime based on factors such as weather, visitation data,orspecialeventsatthesite.

– Weather Conditions: Weather has a direct impact on outdoor activities, transpo-rtation, andconnectivitytoplaces.

– Environmental Sustainability: The growing importance of environmental conservation influencestourdecisions.

• AI & ML:AIandmachinelearningtechnologieshave theabilitytotransformtravelitineraryplanningby allowing for intelligent data analysis, recognizing user preferences, and anticipating travel patterns. Hereareafewsignificantapplications:

Dataminingandknowledgediscovery:Extract relevantinsightsfrommassivetraveldatasets, including feedback from customers, location metadata,andhistorypreferences.

– Anomaly detection: Identify anomalous patterns in journey data, such as unexpected disruptionsorvariationsfromtripplans,and providereal-timenotificationsandremedies.

– Predictive maintenance: Anticipate probable travel- related concerns, such as delays or cancellations,andproposealternativeoptions aheadoftime.

• Support Navigation:Itincludesanavigationsystem to help people find their way around. Sometimes guidancetoaspecificdestinationislost,makingit difficultfortravelerstomaketheirwaythrougha newplace.Toaddressthisissue,anavigationsystem wouldassisttouristsinarrivingprecisely.

2.2 Research Gaps and Future Directions

Despite major advancements in customized trip itinerary planning,variouschallengesremain.

• Limited accuracy of real-time data:Manysystems rely on external sources of information, such as weather, traffic, or alerts, which may not be accurateortimely.

• Insufficient user customization: Many systems demandmorespecificuserinputtoenablegenuine customization.

• Dealing with fake reviews: Fake or biased user reviewscontinuetobeamajorissueinsystemsthat dependonreviewdatafordecision-making.

• Offline accessibility:Fewprogramsprovideoffline capability,whichiscriticalforusersgoingtoremote locationswithlimitedaccess.

Futureresearchdirectionsinclude:

• Improve the algorithms for processing and validating real-time data from different sources, resulting in more accurate and dependable suggestions.

• Providedeeplyindividualizedtravelitinerariesby lever- aging customer behavior analytics, interest clustering,andlong-termdatalearning.

• Include environmentally friendly choices into itineraries, with a focus on sustainable travel practices.

• Create algorithms that promote equitable representation of lesser-known destinations, resultinginvariedandinclusiveitineraries.

2.3 Summary

AI-poweredpersonalizedtravelscheduleplannersrepresent agame-changingapproachtotripplanning,utilizingcuttingedgetechnologiestoprovidepersonalized,efficient,andsustainabletravelexperiences.Thestudyemphasizestheuseof machine learning, reinforcement learning, and algorithms like MCTS, LKH, and deep learning to create dynamic and contextualroutes.Thesesystemsusereal-timedata,suchas weather,ridership,anduserpreferences,toimprovetravel schedulesandhandleissuesincludingovercrowding,slow start times, and customer group travel dynamics. Sustainabilityisamajorpriority,withdesignsthatinclude eco-friendlysolutions,reduceCO2emissions,andhighlight lesser-known features. Despite advances, there are still limitations in offline functioning, scalability, and cultural contextincorporation.Futurepathswillemphasizereal-time adaptation, stakeholder collaboration, and greater personalization in order to offer full, holistic, and usercentrictravelexperiences.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

3. METHODOLOGY AND SYSTEM DESIGN

Togenerateaneffectiveitinerary,Traveltechnologyincorporateshardwareparts,machinelearningmodels,andoptimizationalgorithms.

1) Data Collection and Interpolation:

• Just as sensor nodes gather environmental data,thegadgetcollectsuserdataandchoices from several reassessments (e.g., surveys, previoustravels)toeducategadgetlearningof trends that anticipate the best itinerary possibilities.

• Interpolationintourismcanpredictitineraries for unknown or underrepresented characteristics,suchasadvisingsportsbased onaconsumer’sprofiledespitelimiteddata.

2) Federated Learning Framework:

• In a federated learning system, user data is keptprivatebyprocessingitlocallyandonly transferring version upgrades to a valued server.

• Use federated learning to tailor itineraries basedentirelyonadjacentconsumerdata(e.g., previoustrips,preferences)whilemaintaining privacy.

• UsetoolslikeTensorFlowFederatedorPySyft to provide personalized advice without accessingpersonalinformation,ensuringdata securityandprivacycompliance.

3) Energy-Aware Scheduling:

• The travel route planner can dynamically change the route based on changing factors such as weather, lo- cal events, or user preferenceswhileonthejourney.

• Changetherouteautomaticallybasedonrealtimeparametersincludingweather,available activities,userweariness,andpreferences.

4) Integration of Energy Harvesting:

• To promote travel sustainability, itinerary planners might include ecologically friendly travel options (for example, low-emission vehiclesandsustainablehotels).

Thissystematicmethodologyallowsthetripplannertooperate efficiently and adapt to changing environmental conditions.

3.1 Hardware Infrastructure

ThephysicalinfrastructureforestablishingthePersonalized Travel Itinerary Planner comprises of a number of components that enable data collection, processing, communication,andenergymanagementformonitoringthe environment.Thefollowinghardwarepartsareessentialfor thesystem’sproperoperation:

• Sensor Nodes: The sensor nodes in the route planner can be thought of as the route planner consumer’s device (smartphone, computer) that

gathers data (preferences, location, previous trip history)andinteractswiththecentralsystem.

– Thesmartphone,tablet,orlaptopworksasthe inter-face,collectinguserinputandinteracting withthesystem.

• Microcontroller or Development Board: Thebackendprocessormightbecomparedtoacentralized server or cloud infrastructure that handles route processing and optimization. Recommended optionsinclude:

– Cloudserver: Acentralized server(e.g., cloud ser- vice) that processes user data, generates routes,andrunsmachinelearningmodels.

 Wireless Communication Modules: The communication module guarantees that data is transmitted efficiently between the end user and the centralized system when creating or amending routes. To ensure constant communication, the followingmodulesarerecommended:

– Communication protocol: Uses protocols like MQTTorHTTPtodeliveruserdatatoacentral serverordatabasesforprocessingandupdates.

 Energy Supply (Energy-Efficient Travel): Plannerscanusesustainableenergyoptionstorecommend energy- efficient travel routes and optimize batteryusage.

– Sustainable travel: Recommend energyefficientmodesoftransportation,suchastrains orbusesoverflightsorvehicles.

3.2 System Software and Frameworks

The project’s software infrastructure will include a wide rangeofprogrammingequipment,libraries,andframeworks fordatacollection,processing,andanalysis.Thefollowing software components are necessary for the system to function:

 Programming Languages: Python, C/C++, and frame-workssuchasNode.JavaScriptcanbeused to create backend systems, communication protocols, and device mastering models for itineraryoptimization.Thelanguagesare:

– Python: For information processing, device mastery,andreal-timeitinerarycreation.

– C/C++:C/C++orJavaScriptfordevelopinguser interfaces that collect and display personal information.

 Operating Systems: The server or base station’s operating system should be capable of handling both machine learning activities and communication protocols. The recommended operating systemsinclude:

– Windows-based OS: For the main server or basestation.

 SoftwareLibrariesandFrameworks:Librariessuch as TensorFlow, PyTorch, and Scikit-examine can

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

help teaching models predict and optimize route itineraries based on individual preferences and real-timedata.

– TensorFlow,PyTorch:Tousemachinelearning models for data interpolation and federated learningtechniques.

–

NumPy,Pandas:Fordatamodificationandpreprocessingoperations.

Scikit-learn: Traditional machine learning methodsincluderegression,classification,and clustering.

– OpenCV: If image or video-based data examination is used, such as for pollution detectionormonitoringsystems.

– Federated Learning Frameworks (e.g., TensorFlow Federated, PySyft): To apply privacy-preservingma-chinelearning.

– NSGA-II (Non-dominated Sorting Genetic AlgorithmII):Formulti-objectiveoptimization.

 Communication Protocols: Communications betweenthesensornodesandthecentralserveris necessary. The data is transferred using the followingprotocols:

• MQTT or HTTP: To transmit data between sensornodesandthecentralserver.

 DataVisualizationandReportingTools:Customers maybeabletoviewtheirroutedetailsinrealtime usingvisualizationtoolssuchasGrafanaorPlotly, as well as travel durations, expenses, and environmentalimpact(e.g.,CO2emissions).

• Grafana, Matplotlib, or Plotly: To create realtimedatavisualizationsanddashboards.

 Database: The device may require databases to store user preferences, itineraries, and historical travelinformation.

• MySQL, SQLite, or NoSQL databases: Keep itineraryrecordsandrelatedinformation(e.g., activitykinds,shipmentoptions)instructured orun-structureddatabases.

4. ALGORITHMIC APPROACHES

An AI-powered personalized travel itinerary planner employssophisticatedalgorithmstoprovidedynamic,efficient, and user-centric trip itineraries. The system starts by gatheringuserchoicessuchasdestination,budget,events, andtravelcompanions,aswellascontextualreal-timedata such as weather, crowd levels, and traffic from APIs. Preprocessingnormalizesandenrichesdatawithothersources, suchasenvironmentmetrics.Recommendationsystemsuse content- based and collaborative filtering to match user choicestoattractions,lodgings,andactivities. Optimization algorithms, such as Genetic Algorithms or Monte Carlo Tree Search, create itineraries that strike a balancebetweencustomer happiness,time efficiency,and environmentalfriendliness,whereasreinforcementlearning constantlyadaptsplansinresponsetoreal-timechanges.To

encourageresponsibletourism,sustainabilitycriteriasuch as CO2 emissions reduction are used in conjunction with multi-objectiveoptimization.Thefinalscheduleisdisplayed inuser-friendlyformatssuchasmapsandtimeframes,and post-trip feedback is integrated into machine learning algorithms to improve future recommendations. This adaptable and sustainable strategy ensures a streamlined andtailoredtravelexperience.

• Machine Learning Algorithms for Data InterpolationandAnalysis[1][3][4]:

– ClassificationModels:Usedforcategorizing personoptionsorlocations.

– LinearRegression/PolynomialRegression: Usedforpredictingnumericaloutputs,e.g., experiencefeesorduration.

– K-Means Clustering: For grouping comparable places or sports primarily basedtotallyonfunctionslikepopularity, cost,orpersonratings.

• DeepLearningAlgorithms[9][15]:

– RNN:Processestextualcontentinputsfor itinerary making plans and sentiment evaluation.

– LongShort-TermMemory(LSTM):Predicts time- based elements like site visitors stylesorclimateconditions.

– Latent Dirichlet Allocation: Analyzes personevaluationsorremarkstopickout options.

• FederatedLearningAlgorithms:

– FederatedAveraging(FedAvg):Itincludes schooling neighborhood fashions on persondevices(e.g.,smartphonesornonpublic computers) and aggregating their updates on a primary server with out sharinguncookedrecords.

• RecommendationSystemalgorithms[12][14]:

– CollaborativeFiltering:Suggestslocations and sports primarily based totally on personconductandoptions.

– Content-Based Filtering: Recommends itinerariesorsportsprimarilybasedtotally onparticularfunctionsofbeyondoptions.

• OptimizationAlgorithms:

– MonteCarloTreeSearch(MCTS):Optimize sequences of selections to maximise travellerdelightandfeasibility.

– Multi-Agent Reinforcement Learning (MARL): Op- timize organization tour making plans or control traveller distributionthroughoutplaces.

– Genetic Algorithms (GA): Generate itineraries with the aid of using evolving answersovermorethanonegenerations, enhancing performance in balancing constraints like price range and hobby diversity.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

5. PERFORMANCE METRICS FOR TRAVEL ITINERARY PLANNER AND MACHINE LEARNING ALGORITHM EVALUATION

AI-poweredpersonalizeditineraryplannersareevaluated usingavarietyofimportantperformancefactors,including personalizationaccuracy,customerhappiness,optimization efficiency,sustainability,andcomputingperformance.Personalizationisevaluatedusingcriterialikeasprecisionand recall,andimprovedAIalgorithmsimprovepersonalization. User feedback and Net Promoter Score (NPS) are used to measurecustomersatisfaction,andoptimizationefficiency guaranteesthatthesystembalancesaspectssuchasbudget and time. Sustainability aims to reduce environmental impact by promoting greener travel options. Computing performance, including fast reaction times and efficient resourceutilization,iscrucialforprovidingaconsistentuser experience. These variables work together to build dependable, adaptive, and simple solutions that suit the constantlyshiftingneedsofthetravelindustry.

Table-2:PerformanceMetricsforPersonalizedItinerary

Metric Description Importance Assessment Methods

Net Promoter Score(NPS)

CO2 Emission Reduction Percentage

Used to assess user satisfaction andloyalty

Measuresthe reductionin carbondioxide emissions achieved through sustainable travelchoices

Assesses overalluser satisfaction withroutes generated basedon feedbackand surveys

Assessesthe Envir-onmental andsocioeconomic benefitsof proposedroutes

Segmentation analysis, survey- based questions

quality reducecosts andtime effectivenessof optimization algorithmsin generating efficienttravel plans programming, mixedinteger programming, orheuristic evaluation methods

Scalability Abilitytoscale withoutlossof performance

Waste Reduction

Computational Efficiency

Evaluatethe effectiveness ofthesystem inpromoting sustainable practicesthat reduceenvironmentalwaste,suchas singleuse plastics,food waste,orimproperdisposalwhile traveling

Thetimeand resources requiredfor machine learningtasks

Evaluatesthe efficiencyand scalabilityofa systemasthe numberof usersor destinations increases

Evaluatethe reductionin wastegenerationfrom recommendati onsfroman AI-based travelitinerary planner

Increasing the numberof concurrent usersor requests determines thesystem capacity

Identifyspecificsourcesof waste,calculatetheaverage waste generatedby common traveloptions, suchasunsustainable accommodatio ns,meals,and activities

Evaluatethe computational resources requiredby thealgorithm

Latency Referstothe timeittakes forthesystem toprocessa userrequest andprovide anoutputor response

Evaluatesthe efficiencyand scalabilityof thesystemas thenumberof usersor destinations increases

Usestandard Fo-rmulas (e.g. DEFRAOrIPCC emission factors) to calculate emissions fordiff erenttravel optionsbasedon dis-tance,mode oftransport and ene-rgy efficiency

Latency requiresrealtimeevent updatesand responses

Solution Abilityto Measuresthe Linear

6. CONCLUSION

Measuretime foreachtask, evaluate hardware/soft ware requirements

Thisextensionaimstoimprovesurgeexpectationandhazardevaluationbyintegratingmachinelearning(ML)models, including Arbitrary Woodlands (RF), which have been shown to be effective in several scenarios. We will significantly improve the accuracy of surge chance predictionsbyleveragingavarietyofinformationsources, including geomorphic, socio-economic, and flexibility components. The survey findings show the success of ML models in spike risk mapping, harm evaluation, and early warningsystems.IrregularWoodlands,inparticular,have demonstratedsuperiorperformanceacrossseveraldatasets, makingthemaneffectivetoolforsurgechancemanagement. TheTravelItineraryGeneratorhasalonghistory,focusing onincreasingclientinvolvementandusefulness.Theextent canbecoordinatedwith:

• Integration with Wearable Gadgets:Enablerealtimerouteandnotificationalertsonsmartwatches orfitnesstrackerstoimprovetripcomfort.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

• Multilingual Bolster:Consolidatemultipledialects to appeal to a global audience, ensuring convenienceandaccessforinternationaltourists.

• Integration with AR/VR Innovation: Utilize augmented reality (AR) to provide immersive experiences,suchasvirtualglimpsesatgoalsorARguidedtours.

REFERENCES

[1] Ki-Beom Kang1 Myeong Gyun Kang2 Seong-Hyuk Jo2 and Jeong- Woo Jwa3 1 National League of Fisheries Cooperatives,Jeju,Korea2Student,3Professor,Division ofMedia transmissionEng., Jeju National Univ., Korea email:kkb8671@gmail.com,2015108101,polim0209, lcr02@jejunu.ac.kr.

[2] AnkitaMudhale1,MadhuriShirmale2,VedantKudalkar3, RishikeshMotiray4,ShariqueAhmad51,2,3,4Computer Engineeringstudent,Uni-versalcollegeofEngineering 5AssistantProfessor,UniversalcollegeofEngineering.

[3] R.Regin*,S.SumanRajest1.DepartmentofComputer ScienceandEngineering,SRMInstituteofScienceand Technology, Ramapuram, India. 2. Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India * Correspondence:regin12006@yahoo.co.id.

[4] AiliChen,XuyangGe,ZiquanFu,YanghuaXiao,Jiangjie Chen Fudan University System Inc. alchen20, xyge20, shawyh,jjchen19@fudan.edu.cnfrank@system.com.

[5] R.K.Sahoo,S.Sethi,S.K.UdgataDepartmentCSEA,IGIT Sarang, Dhenkanal, India School of Computer and Information Sciences, Uni- versity of Hyderabad, Hyderabad,IndiaTheAuthor(s),underexclusivelicense toSpringerNatureSingaporePteLtd.2021Intelligent Systems,LectureNotesinNetworksandSystems.

[6] Sparsh,AbhishekKumarJadaun,ArjunPandit,JabirAli, Mohd Shariq School of Engineering andT echnology, Department of Computer Sci- ence and Applications ShardaUniversity,GreaterNoida,UP,India.

[7] ArisMunandar*,HanifFakhrurroja,MuhammadIlham Rizqyawan,RianPutraPratama,JonyWinaryoWibowo, IrfanAsfyFakhryAntoTechnicalImplementationUnit forInstrumentationDevelopment(UPTBPI)Indonesian InstituteofSciences(LIPI)Bandung,Indonesia

*aris001@lipi.go.id

[8] PersonalizedTourItineraryRecommendationAlgorithm BasedonTouristComprehensiveSatisfactionDingming Liu1,*,LizhengWang1,YanlingZhong1,*,YiDong2,3 andJinlingKong1,1SchoolofGeologicalEngineering and Geomatics, Chang’an University, Xi’an 710054, China 2 Aerial Photogrammetry and Remote Sensing

Group Co., Ltd., Xi’an 710199, China, 3 Shaanxi Engineering Research Cen- ter of Geospatial Information, Xi’an 710199, China Correspondence: dmliu0401@gmail.com(D.L.);2019026021@chd.edu.cn (Y.Z.).

[9] A survey on personalized itinerary recommendation: FromoptimisationtodeeplearningSajalHaldera,Kwan HuiLimb,JeffreyChana,Xi-uzhenZhangaaSchoolof ComputingTechnologies,RMITUniversity,Australiab Singapore University of Technology and Design, Singapore.

[10] Optimizing Travel Itineraries with AI Algorithms in a Microservices Architecture: Balancing Cost, Time, Preferences,andSustainabilityBimanBaruaa,b[00000001-5519-6491]andM.ShamimKaiserb,[0000-00024604-5461]aDepartmentofCSE,BGMEAUniversitsyof Fashion&Tecnnology,Nishatnagar,Turag,Dhaka-1230, Bangladesh b Institute of Information Technology, Jahangirnagar University, Savar- 1342, Dhaka, Bangladeshbiman@buft.edu.bd.

[11] Multi-objective sustainability tourist trip design: An innovativeapproachforbalancingtourists’preferences with key sustainability considerations Rapeepan Pitakaso a , Thanatkij Srichok a , Surajet Khonjun a , Sarayut Gonwirat b Natthapong Nanthasamroeng c, Chawis Boonmee d,e, a Artificial Intelligence Optimization SMART Laboratory, Department of Industrial Engineering, Faculty of Engineering, Ubon Ratchathani University, Thailand b Department of Computer Engineering and Au- tomation, Faculty of Engineering and Industrial Technology, Kalasin University, Thailand c Department of Engineering Technology, Faculty of Industrial Technology, Ubon Ratchathani Rajabhat University, Thai- land d Department of Industrial Engineering, Faculty of Engineering, Chiang Mai University, Thailand e AdvancedTechnologyandInnova-tionManagementfor Creative Economy Research Group, Chiang Mai University,Thailand.

[12] Personalized Tourist Recommender System: A DataDriven and Machine-Learning Approach Deepanjal Shrestha 1,2, Tan Wenan 1 , Deepmala Shrestha 2, NeeshaRajkarnikar2andSeung-RyulJeong3,1School ofComputerScienceandTechnology,NanjingUniversity ofAeronautics&Astronautics,Nanjing211106,China; wtan@foxmail.com, 2 School of Business, Pokhara University,Pokhara33700,Nepal,3GraduateSchoolof BusinessIT,KookminUniversity,Seoul02707,Republic ofKorea.

[13] AnalysingTouristExperiencesinResponsetoAI-Based DigitalTech-nologiesAdaption:ALogisticRegression AnalysisinCaseofUzbek-istanGurinderSinghAmity University Noida, Naina Chaudhary Amity University,

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01| Jan 2025 www.irjet.net p-ISSN: 2395-0072

Tashkent Danish Ather Amity University Tashkent https://orcid.org/0000-0003-1596-5553RajneeshKler Amity University Tashkent, https://orcid.org/00000001-7402-9330, Manik Arora Amity University Tashkent.

[14] RPMTD:ARoutePlanningModelWithConsiderationof Tourists’ Distribution YUNTAO KONG 1 , KUN YI 2 , LIJUN WANG 1 (Graduate Student Member, IEEE), CHENGPENG1,LE-MINHNGUYEN1,ANDQIANGMA3 ,(SeniorMember,IEEE)1SchoolofInformationScience, JapanAdvancedInstituteofScienceandTechnol-ogy, Nomi,Ishikawa923-1292,Japan2InstituteofEconomic Research, Kyoto University, Kyoto 606-8501, Japan 3 DepartmentofInformation Science,KyotoInstituteof Technology,Kyoto606-0951,Japan.

[15] Promoting sustainable tourism by recommending sequences of attrac- tions with deep reinforcement learningAnnaDallaVecchia1·SaraMigliorini1·Elisa Quintarelli1 ·Mauro Gambini1 · Alberto Belussi1 Received:26September2023/Revised:1March2024/ Accepted:25March2024/Publishedonline:24April 2024©TheAuthor(s)2024. M.Young, The Technical Writer’sHandbook.MillValley,CA:UniversityScience, 1989. R. Nicole, “Title of paper with only first word capitalized,”J.NameStand.Abbrev.,inpress.K.Elissa, “Titleofpaperifknown,”unpublished.

2025, IRJET | Impact Factor value: 8.315 |