Issuu

A Review on Dharaansh: The Sustainable Farming Advisor using Machine Learning

M. D. CHOUDHARI1 , KHUSHI DHABALE2 , EKTA SATGHARE3 , DIYA TEMBHURNE4 , SAHMITA , NINAVE5 .

1Assistant Professor AI & DS Department K. D. K. COLLEGE OF ENGINEERING, Nagpur

2,3,4,5 K. D. K. COLLEGE OF ENGINEERING, Nagpur

ABSTRACT- Dharaansh: A Sustainable Farming Advisor based on machine learning can significantly enhance crop productivity and resource use efficiency while providing site-specific data-driven recommendations to the farmer. The system will integrate real-time and historical data on soil pH, soil temperature, and moisture and key climatic variables like ambient temperature, humidity, and rainfall to produce advisory outputs on suitable crops, alternate cropping options, and fertilizer management strategies. This model is trained using supervised learning algorithms on curated agronomic datasets that comprise soil properties, nutrient levels, and weather patterns to predict the most suitable crops and optimal fertilizer regimes for given field conditions, minimizing the risk of crop failure and overuse of inputs. The system also uses rule-based logic and sensor thresholds that raise flags for highly acidic or alkaline soils, suggesting corrective soil amendments like lime or organic matter to restore soil health. The application combines intelligent recommender techniques to make personalized, context-aware recommendations to support precision agriculture and sustainable farming practices. Overall, this Sustainable Farming Advisor seeks to enable efficient fertilizer application, resilient crop planning in response to changing climate conditions, and accessible real-time decision support for farmers.

Keywords: Machine learning, Sustainable farming advisor, soil Ph, Crop recommendation, Smart farming, Sustainable farming, Fertilizer recommendation.

INTRODUCTION

Farmingisthebackbone ofIndia'seconomy,but millionsoffarmerscontinue tograpple withunforeseeableweather, lesseningsoilfertility,pestattacks,anduncertaintyduetoalackofdependableinformation.Toempowerfarmersonsuchcriticalissues,anAI-poweredagriculturalsupportsystembythenameDharaanshhasbeendeveloped,knownasDharaansh:The SustainableFarmingAdvisor.

DharaanshfocusesonaccuratecroppredictionbyanalyzingsoilpropertiessuchaspHandNPKlevelsalongwithatmospheric conditionsoftemperature,humidity,andrainfalltorecommendthemostsuitable cropsandreducetheriskofcropfailure.It also focuses on soil health and fertility management by providing pH validation, fertility checks, and corrective suggestions thatincludeorganicamendments,biofertilizers,andcroprotation.

Itgivesreal-timeweatheradvisoryservicesforguidanceonschedulingirrigation,harvest,andmulti-croppingactivitieswhile safeguardingagainstadverseweatherconditions,enablingsustainableandprofitablefarming becauseofpredictivepestand diseasealertsintegratedintothesystem,besidespromotionofeco-friendlyfarmingpractices.

Withitsregionallanguagesupport,awarenessofgovernment schemesandauser-friendlyinterfacepresentingvisualreports and insights, Dharaansh helps bridge the gap between traditional farming and modern technology, empowering farmers to achieveproductivity,profitability,andlong-termsustainability.

LITERATURE REVIEW

Most recent work on sustainable crop and fertilizer advisory systems focuses on software-centric, machine-learning modelsoperatingoncurateddatasetsratherthanliveIoTstreams.TheseusuallyingestsoiltestreportsofpH,N–P–K,organic carbon,historicalweatherdata,andcropperformancerecordsstoredinrelationaldatabasesorCSVfiles;thenapplyclassification or ensemble algorithms that generate recommendations. Several studies propose web-based or desktop applications where farmers or extension workers would enter soil and climate attributes manually, and the backend ML engine returns

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

suitablecropsandfertilizerplans,thusprovingthatevenofflineorlow-connectivitysettingscanbenefitfromAI-drivendecisionsupport.

The classical machine-learning techniques dominate the software-based literature. Many experiments compare the performanceofalgorithmssuchasK-NearestNeighbors,DecisionTrees,NaïveBayes,SupportVectorMachines,LogisticRegression, Random Forest, XGBoost, and Gradient Boosting on benchmark crop recommendation datasets, often recording accuracies above 95% for the best models. Some works extend the basic task of crop prediction to recommending fertilizers through learning mappings fromsoil nutrients and target croptotype and quantity of fertilizer, mostly using multi-outputclassification or regression frameworks. More recently, works have also focused on explainable AI, where feature-importance scores andinterpretationmethodshelpagronomistsandfarmersunderstandwhyaparticularcrop-fertilizercombinationisrecommended.

Another important strand of software-based research emphasizes integrated advisory platforms, which integrate crop recommendation,fertilizeroptimization,andattimesevenpriceormarketguidanceintooneAI-poweredportal.Thesesystems provide user-friendly dashboards and form-based input screens, allowing non-technical users access to advanced analytics withouthavingtohandlerawcodeorconfigurations.Reviewsofsuchplatformsarguethatpuresoftwaresolutions,trainedon high-qualitysoilandclimatedatasets,cansignificantlyimproveinput-useefficiencyandyieldoutcomesevenpriortofullIoT integration,especiallyinregionswithdevelopingsensorinfrastructure.Thisliteratureprovidesessentialgroundingfora machine-learning-based,software-drivenSustainableFarmingAdvisorgroundedonsoilpH,temperature,humidity,andclimatic datainrecommendingsuitablecrops,alternates,andfertilizers.

COMPARATIVE ANALYSIS & RESEARCH GAP

Study Methodology

Shastrietal., 2025 GradientBoostingEnsembleModel

Soil nutrients, pH, climaticattributes

Baishya,2025 (TinyML) TinyML and LightweightRandomForest.

IoT-based real-time sensing, on-device inference

Improved general and allinclusive decision-making rate and dependability using enhanced feature inputs.

Sam&D’Abreo, 2025 Random Forest, Support Vector Machine and Temporal Evaluation.

Prity, 2024 (Springer)

Random Forest, Support Vector Machine,

Environmental and Economic parameters

Yield history with farmerpreferences

Accomplish faster, lowpower forecasts suitable for edge positioning or placement.

The study suffersfrom limitedgeographic difference in the dataset, which limits themodel’scapacitytoextrapolateacross different agricultural -climatic portions. Also,theabsenceoftemporalorseasonal validation elevates challenges of overfitting,asthemodelmaynotaltereffectivelytoyearlyclimatevariations.Thelackof transparencyfurtherreducesitspractical applicability for farmers who need lucid recommendations.

In spite of strong TinyML performance, the model has yet to be authenticated under real field conditions where sensor noise, environmental variations, and hardware inconsistencies can affect dependability.Itsdatasetremainsrelatively small, restricting model transferability to different soil types. Moreover, long-term aspectssuchasdeviceageing,calibration drift, and durability of on-device inferencearenotexamined.

Offers more practical performance through timeaware testing and better perceptions for farm plannings.

Dependence on contrast to economic data, it restricts hardiness, and the absenceofadaptivelearninglimitations,the model’s ability to respond to changing climaticpatterns.

Provides more purified recommendations as Use of open datasets minimizes local significance,andthelackoftransparency

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

Kiran, 2024 (MDPI)

Gopi, 2023 (HMFO-ML)

Artificial Neural Network

End-to-end ML System (Random forest/Gradient Boosting Machine)

HMFO + Probabilistic Neural Network + Extreme Learning Machine

COMPARATIVE SUMMARY

Multi-source integration with user interface

Merged or Unified crop recommendation with yield forecasting

compared to algorithmic systems. limits acceptance. Field deployment and real-worldevaluationsaremissing.

Strengthens decision support by combining soil and weather data into a centralizedplatform

Enhances system performance through composite optimization and bifunctional-taskmodeling

Location biasin data minimizes transferability, and the model lacks clearness. Real-time weather integration is also missing.

High algorithmic complexity limits practicaldeployment,andtheabsenceofgeographic or multi-region testing affects abstraction. Contrasts with newer DL methodsortechniquesarealsomissing.

A relative study of DHARAANSH with the seven existing crop recommendation systems indicates that your model is uniqueforitsadaptability,interpretability,anddesignforfield-leveldeployment.Whilemostrelatedworkinrecentliterature focusesoneitheralgorithmicaccuracyorcomputationalenhancements,DHARAANSHfocusesonregionalscalability,real-time environmental responsiveness, and farmer usability, which have not been well addressed in most current literature. For example,theGradientBoostingmodeldevelopedbyShastrietal.(2025)achieveshighcomputationalefficiencybutreliesheavilyonageographicallynarrowdataset,therebyseverelylimitinggeneralizationtootheragro-climaticzones;DHARAANSHresolvesthisthroughfacilitatingflexibilityindatasetexpansionandusingafeaturedesignconsideringlocalizedsoilvariations andvariouscroppingpatterns.Similarly,TinyML-basedsystemssuchasBaishya(2025)achieveoveralllow-power,on-device inference but their validity is yet to be determined under real agriculture noise conditions, for example, fluctuating sensor quality, moisture interference, or seasonal drifts. By contrast, DHARAANSH is structured to integrate stable pre-processing andnoise-handlingmechanismstogiveitfieldrobustness.

Combined environmental and economic parameter models, such as Sam & D'Abreo 2025, have wider analytical capabilities but are still prone to possible unreliability in economic data and the absence of adaptiveness in their learning framework. DHARAANSHovercomessuchweaknessesbyfacilitatingregularupdatesofdatasetsandallowingthemodelstochangewith changingclimaticoragronomicconditions.TheapproachesdescribedbyPrity2024andKiran2024havetheirfocusonuserfriendlinesswithmulti-sourceintegration,buttheylacktestingatthedeploymentlevelanddonotprovideclaritywithrespect totheuncertainty of the prediction-an extremely important factor in decision-making relatedtofarming.DHARAANSHcompensates for this by placing a strong emphasis on interpretability, integrating transparent feature-impact analysis, thereby makingtherecommendationsmoreunderstandableandcrediblefortheendusers.

The Rao et al. (2024) ensemble frameworks enhance stability and reduce the variance in predictions. However, this is constrained by the geographically biased datasets the assembling relies on. DHARAANSH has been designed to be modular and region-adaptive; it can embed region-specific calibration to avoid these limitations. Likewise, metaheuristic-based models suchasGopi2023showstrongoptimizationbutarecomputationallyheavytotheextentthatmosttypicalruralsettingshave verylimitedresourcestoaccommodatethem.DHARAANSHavoidssuchhighcomplexitybyemployinglightweightyetpowerfulmodelswithreal-worlddeploymentfeasibilityinmind.

METHODOLOGY

1. Data Acquisition from User

Theuserinputsnecessaryinformationregardingtheenvironmentandsoilthroughthesystem.Thecollectedinputparametersare:

 Temperature

 Humidity

 SoilType

 SoilpH

 Rainfall

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

 Region

 Season

Theseinputsactasthefoundationforallfurtherpredictionsandrecommendations.

2. Input Processing

Theinputvaluesaretakenbythesystemandvalidated.Itchecks:

 Whetherallrequiredfieldsarefilled

 Whetherthevaluesfallwithinacceptableagriculturalranges

Thevalidateddataisthenpassedtotwomodules:

 CropRecommendationEngine

 CropHistoryCheckModule

3. Crop Recommendation Engine

Thesystemcomparesthevaluesofenvironmentalandsoilparameterswithagriculturaldatasets.Themodulehighlights:

 PrimaryRecommendedCrop

 AlternativeCropOptions

Thesystemfurthercalculatesapproximatelevelsof:

 Nitrogen(N)Requirement

 Phosphorus(P)Requirement

 PotassiumRequirement

Themodulealsopredictsthesuitableirrigationmethod,dependingon:

 Rainfall

 Soilmoisturecapacity

 Cropselection

Theoutputsofthismodulearesentfordisplaytotheuser.

4. Crop History Check Module

Themoduleagainreceivesthesameinputs,especiallySoilType,Region,andSeason.Itretrievesthehistoricalcroppattern oftheland.Itcheckswhetherthesamecrophasbeengrownrepeatedly:

 Ifthesamelandhasbeenusedforthesamecrop3ormoretimesawarningisgenerated. Outputsfromthismoduleinclude:

 Repeated-cropwarning

 Soilfertilitywarning

 Suggestionsforrestoringsoilfertility:croprotation,organicmanure,greencovercrops,etc.

 Theresultsarepassedontotheuserinterface.

5. Soil pH Monitoring and Alert System

ThesystemwillcheckwhethersoilpH<5.5,indicatingacidicsoil.

2.IfpHisbelowthethreshold:

 ApHwarningmessageisgenerated.

 RecommendationsforimprovingsoilpHaredeveloped(e.g.,lime,organiccompost,gypsumaddition).

6. Output Generation

Thesystemcompilesfinalresultfromallthemodulesanddelivers:

1. CropRecommendation

 RecommendedCrop

 AlternativeCrop

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

 RequiredNPKlevels

 Irrigationtypesthataresuitablefor

2. Warning&Alerts

 Soilfertilitywarning

 Croprepetitionwarning

 SoilPHwarningifPH<5.5

3.ImprovementSuggestions

Soilrestorationmethods

 soilPHimprovementsuggestion

7. System Feedback Loop

OptionalFutureEnhancementtoachievegreateraccuracyovertime:

 Theuserfeedbackregardingcropperformancecanbestored.

 Refinerecommendationsbytrainingmachinelearningmodels.

PROPOSED WORK

1. User Interaction Layer

Thesystemstartsbyhavingtheuserinputenvironmentalandsoilinformationliketemperature,humidity,pH,rainfall,and type of soil, region, and season of cultivation. This information is gathered through the frontend interface and sent to the backendprocessingunit.Thesystemcheckstheinputformatforcorrectnessandvalueswithintheiracceptablerange.

2. Data Processing Layer

Afterreceivinginput,thesystempre-processesthedata:

 Convertsrawvaluestostandardizedunits

 Itclassifiessoiltypeandseason

 Checksformissingorinconsistentdata

Thecleaneddataisthenpassedontoitsrespectivemodulesforanalysis.

3. Crop Recommendation Module

Thesystemcomparesthereceivedparameterswithagriculturalknowledgedatasets. Itfindsthebestcropsuitedforthegivenlandbasedonthematchingconditionsoftemperature,soiltype,rainfall,region,and soon.Itgenerates:

 RecommendedCrop

 AlternativeCrop(s)incasetheprimarycropisrisky Thismodulealsopredicts:

 IdealNPKfertilizerrequirements(Nitrogen,Phosphorus,Potassium)

 Suitableirrigationmethod:drip,sprinkler,floodirrigation

4. Crop History Verification Module

Previouscroppingcyclesofthesamelandarechecked.Itchecksifthesamecrophadbeengrownmorethanthreeconsecutivetimes.Whenrepeatedcroppingisdetected:

 EMSgeneratesawarningaboutsoilfertility.

5. The system also provides:

Croprotationrecommendations.Soilrestorationtips:organicmanure,greenmanure,microbialculture,etc.

6. Soil pH Monitoring Module

Themoduleanalyzesthesoil pHinput.IfthecalculatedpHislessthan5.5,thefollowingisautomaticallyinitiated:

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

 Asoilaciditywarning

 RecommendationsonimprovementinsoilpH(applicationoflime,compost,ash,etc.)

7. Output Generation Layer

Afterreceivingresultsfromallmodules,thesystemcompilesthemintoauser-friendlyoutput.Finaloutputis:

 Recommendedcrop

 Alternativecrop(s)

 N,P,Kfertilizerlevels

 Irrigationmethod

 SoilpHwarning

 Soilfertilitywarning

 Repeatedcropwarning

 Suggestions

 PHimprovementstrategiesforsoilfertilityrestoration

8. Decision Feedback (Optional Enhancement)

Theendmaygivefeedbackafter adherringtosuggestions.2.Futureversionsshouldincludemachinelearningmodelsto increaseaccuracybyusingpreviousoutcomes.

PROJECTED OUTCOME AND FUTURE SCOPE

TheDhaaranshsystemwillofferreliablecropsuggestionsthroughanalysisofsoilandenvironmentalfactors.Inreturn, itwillhelpfarmersintheselectionoftherightcropwhileenhancingyieldandimprovinguseofresources.Theplatformwill alsocreateawarenessforsoilhealthbyprovidingwarningsrelatedtopH,fertilityalerts,andbalancedNPKrecommendations Thiscanreducesoildegradationandenhancelong-termlandsustainability.Byguidingfarmersonpropermethodsofirrigationandfertilizerrequirement,thesystemintendstoreduceunnecessaryexpenditure.

Inthe future, Dhaaransh can be designed to becomea more enhanced smart farmingsolution. This will be achieved by integratingreal-timeIoTsensors,machinelearning-basedpredictionmodels,andweatherforecastAPIsforbetteraccuracy.Additionalfeaturesincludethepredictionofpestsanddiseases,integrationofasoilhealthcard,andfarm-to-marketpriceforecasting,whichshallfurtherassistfarmers.

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

Submit

Inputs

Temperature

Humidity

SoilType

Rainfall Region Season

User

Inputs

Temperature

Humidity

SoilType

Rainfall Region Season

Outputs receives

RecommendedCrop

AlternativeCrop

Nitrogen

Phosphorus

Potassium

IrrigationType

Receives Triggers

Triggers triggers

Crop History Check

Sameland≥3times

SoilfertilityWarning

Soilfertility

RestorationSuggestions

Soil pH<5.5

SoilpHwarning

SuggestionstoimprovepH

Thesystemshallalsotransformintoamultilingualmobileappbyincorporatingsmartirrigationautomationandalertswithin thefarmingcommunityforpestorclimatehazards.ThesechangeswillmakeDhaaranshafull-fledgeddigitalagricultureassistant,promotingsustainablefarmingandincreasingproductivity

CONCLUSION

DharaanshaimstobeanintegratedAI-enabledagriculturesupportsystemthatassistsfarmerswithinformeddecisionsfor sustainableandprofitablefarming.Basedonsoil,weather,pest,andmarkettrend analysis,itundertakescrop recommendations, timely alerts, and expert advice. The potential to redefine traditional farming practices is still quite high though many obstacleslieahead,fromdatainaccuracyandconnectivitytohurdlesintechnologyadoption.AdvancedAImodelsandmarket transparencytoolscancontributemuchtowardsenhancingcropyield,minimizinglosses,andpromotingsustainableagricul-

Fig 1. Flowchart

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

ture. Overall, Dharaansh represents a step toward smart farming, empowering farmers with technology to secure better productivityandeconomicgrowth.

REFERENCES

[1] Yu, S. & Mu, Y. (2022). Sustainable Agricultural Development Assessment: A Comprehensive Review and Bibliometric AnalysisDOI:https://doi.org/10.3390/su141911824

[2] Yi Ding,Luo Jun,Gao Haodic,Zhang Xing,Ye Lie,Siti Sarah Maidin,Wan Hussain Wan Ishak,Walton Wider(2024). TransformingAgriculture:AnInsightintoDecisionSupportSystemsinPrecisionFarming.JournalofAppliedDataSciences DOI:https://doi.org/10.47738/jads.v5i4.274

[3] Alebachew Chiche(2024).HybridDecisionSupportSystemFrameworkforCropYieldPredictionandRecommendation. InternationalJournalofComputing.DOI:https://doi.org/10.47839/ijc.18.2.1416

[4] T. Jalaja, T. Adilakshmi, Meka Dheeraj Reddy, K. Sai Kamal (2024). Empowering Agriculture: A Machine Learning-Based Decision Support System for Crop Selection and Profitability Analysis. International Journal of Intelligent Systems and ApplicationsinEngineering.DOI:https://ijisae.org/index.php/IJISAE/article/view/5800

[5] Muhammad Indra Gunawan,Joni Wilson Sitopu,Gamar Sechan,Indra Gunawan(yearN/A).OptimizingCropSelection:A Multi-CriteriaDecisionSupportSystemforSustainableAgriculture.InternationalJournalofEnterpriseModelling. Link:https://ieia.ristek.or.id/index.php/ieia/article/view/94

[6] Martine J. Barons,Lael E. Walsh,Edward E. Salakpi,Linda Nichols(2024).ADecisionSupportSystemforSustainableAgricultureandFoodLossReductionunderUncertainAgriculturalPolicyFrameworks.Agriculture,14(3),458. Link:https://doi.org/10.3390/agriculture14030458

[7] Meenakshi Malik, Mukesh Sehgal, A.K. Kanojia, R.V. Singh (2018). A review paper on decision support system/expert systemdevelopedonmango.InternationalJournalofPlantProtection,11(1),119-123.arXiv: Link:https://doi.org/10.15740/HAS/IJPP/11.1/119-123

[8] Zhai Z.,Martínez J.F.,Beltrán V.,Martínez N.L.(2020).Decisionsupportsystemsforagriculture4.0:Surveyandchallenges.ComputersandElectronicsinAgriculture,170,105256. DOI:https://doi.org/10.1016/j.compag.2020.105256

[9] Iffat Ara, Lydia Turner, Matthew Tom Harrison, Marta Monjardino, Peter de Voil, Daniel Rodriguez (2021). Application, adoptionandopportunitiesforimprovingdecisionsupportsystemsinirrigatedagriculture:Areview. AgriculturalWaterManagement,257,107271Link:https://ideas.repec.org/a/eee/agiwat/v257y2021ics0378377421004388.html

[10] S Sridevy, M Nirmala Devi, M Sankar (2023). Decision support systems in agricultural industry perspective. InternationalJournalofStatisticalandAppliedMathematics,8(2S):29-31. DOI:https://doi.org/10.22271/maths.2023.v8.i2Sa.984

[11] Desta Abayechaw (2021). Review on Decision Support System for Agrotechnology Transfer (DSSAT) Model. InternationalJournalofIntelligentInformationSystems,10(6),117-124.DOI:https://doi.org/10.11648/j.ijiis.20211006.13

[12] John E. Ikerd (1991). A Decision Support System for Sustainable Farming. Northeastern Journal of Agricultural and ResourceEconomics,20(1),109-113DOI:https://doi.org/10.1017/S0899367X00002919

[13] Luiz Henrique A. Rodrigues, Fabio Zimmermann (2025). Digital Decision Support Systems for Sustainable Organic Farming.InformaticsStudies,12(2).Link:https://informaticsstudies.org/index.php/informatics/article/view/741

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

Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072

[14] Emmanuel Asolo,Ifechukwu Gil-Ozoudeh,Chibundom Ejimuda(2024).AI-PoweredDecisionSupportSystemsforSustainableAgricultureusingAI-ChatbotSolution.JournalofDigitalFood,Energy&WaterSystems,5(1). DOI:https://journals.uj.ac.za/index.php/DigitalFoodEnergy_WaterSystems/article/view/3260

[15] Banashri Rath,DecisionSupportSystemforDraftingofSustainableAgricultureProductionPolicyforOdisha. InternationalJournalofSupplyChainManagement,2(2).DOI:https://doi.org/10.59160/ijscm.v2i2.745

[16] Vidya G. Kumbhar,T.P. Singh(2013).ASystematicStudyofApplicationofSpatialDecisionSupportSysteminAgriculture.InternationalJournalofScientific&EngineeringResearch,4(12). Link:https://www.ijser.org/researchpaper/A-Systematic-Study-of-Application-of-Spatial-Decision-Support-Systemin-Agriculture.pdf

[17] Suman Kumar Das, Pujyasmita Nayak (2024/2025). Integration of IoT-AI powered local weather forecasting: A Game-ChangerforAgriculture.arXivpreprint(2024).DOI:https://arxiv.org/abs/2501.14754

[18] Anat Goldstein,Lior Fink,Gilad Ravid(2019).AFrameworkforEvaluatingAgriculturalOntologies.arXivpreprint. arXiv:https://arxiv.org/abs/1906.10450

[19] Suman R. Venu, R. G. Kumar, M. K. Kumar, T. G. Prasad, B. Suresh, P. Neelima (2023). An Intelligent and Service Based Smart Agriculture Recommendation System. International Journal of Intelligent Systemsand ApplicationsinEngineering,12(7s),153-158. Link:https://ijisae.org/index.php/IJISAE/article/view/4051

[20] Digitalagro-advisorytoolsintheglobalsouth:a behavioralanalysisofimpacts,andfuturedirections. Discover Agriculture,3,article 40.(2025).LinkI:https://link.springer.com/article/10.1007/s44279-025-00190-y

2026, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page4