International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072
Agri Invest AI “Intelligent Agricultural Investment, Crop Analytics, and Collaborative Farming Platform Using Artificial Intelligence”
Thrisha J C, Venu P K, Thrisha S, Akshay Kumar A, Chandana S.N
¹Student, Dept. of CSE, CMR University, Bengaluru, India
²Student, Dept. of CSE, CMR University, Bengaluru, India
³Student, Dept. of CSE, CMR University, Bengaluru, India
⁴Student, Dept. of CSE, CMR University, Bengaluru, India
5Professor, Dept. of CSE Engineering, CMR University, Bengaluru, India
Abstract - Agriculture plays a vital role in economic development, yet farmers face numerous challenges such as unpredictableclimateconditions,lowproductivity,andlackof financialsupport.Simultaneously,investorsseekreliableand sustainable investment opportunities but lack access to transparent agricultural platforms. To address these issues, this paper presents Agri-Invest AI, a hybrid system that integratesArtificialIntelligence(AI),FullStackDevelopment, and Financial Technology (FinTech) to create a unified ecosystem for farmers and investors.The proposed system includes a Crop AI Module that predicts crop yield, detects plant diseases using image processing, and recommends fertilizers based on soil and weather conditions. The Investment Module utilizes ESG (Environmental, Social, and Governance) principles to provide intelligent portfolio recommendations and Systematic Investment Plan (SIP) options. A unique Bridge Feature connects farmers and investors,enablingprojectfundingandensuringtransparency. The system is developed using ReactJS for frontend, Spring Boot for backend, MongoDB for database management, and TensorFlow for AI integration. Experimental results demonstrate improved crop productivity, reduced financial risk,andenhancedinvestmentdecision-making.Theplatform promotessustainableagricultureandresponsibleinvestments, contributing to long-term economic and environmental benefits
KEYWORDS (Artificial Intelligence, Crop Prediction, ESG Investment, Sustainable Agriculture, Machine Learning,FinTech,DeepLearning,SmartFarming)
1.INTRODUCTION
Agricultureisoneofthemostcriticalsectorsindeveloping economieslikeIndia,contributingsignificantlytoGDPand employment. However, farmers face persistent challenges suchasclimatevariability,pestattacks,soildegradation,and
lackofaccesstofinancialresources.Thesechallengesresult inreducedproductivityandfinancialinstability.
Ontheotherhand,investorsareincreasinglyinterestedin sustainableandethicalinvestmentopportunities.However, thelackoftransparency,riskassessmenttools,andreliable platforms limits their ability to invest in agriculture. To bridgethisgap, Agri-Invest AI isproposedasanintegrated platform that combines Artificial Intelligence, agriculture, and financial systems. The platform enables farmers to optimize crop production and allows investors to make informed,sustainableinvestmentdecisions.
1.1 Problem Statement
Farmerslackaccesstointelligentcropanalysistools
Difficultyinobtainingfinancialsupport
Investorslacktransparentagriculturalinvestment platforms
High risk due to unpredictable environmental conditions
1.2 Objectives
PredictcropyieldanddetectdiseasesusingAI
ProvideESG-basedinvestmentrecommendations
Connect farmers and investors through a digital platform
Reduceriskandimproveprofitability
Promotesustainableagriculture
2. SYSTEM DESIGN AND METHODOLOGY
The proposed system is designed using a modular architecture that integrates data processing, machine learning,andfull-stackdeployment.Theworkflowisdivided into multiple phases to ensure scalability, efficiency, and accuracy.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072
2.1 Data Collection
Thesystemcollectsmultipledatasetsrequiredforaccurate prediction and analysis. These include soil parameters, weatherconditions,crophistory,andplantdiseaseimages. Thisdataformsthefoundationfortrainingmachinelearning models.
2.2 Data Preprocessing
Rawdataisprocessedtoimprovequalityandconsistency. Thisincludesnormalizationofsoilvalues,handlingmissing data, and image preprocessing such as resizing and augmentation. Feature engineering is also applied to enhancemodelperformance.
2.3 Model Training
Machinelearningmodelsaretrainedusingprocesseddata. Regressionmodelsareusedforcropyieldprediction,while ConvolutionalNeuralNetworks(CNN)areusedfordisease detection.ClassificationalgorithmssuchasRandomForest andSupportVectorMachinesareusedforcropsuitability analysis.
2.4 Deployment
The trained models are integrated into the system using REST APIs. The results are displayed through interactive dashboards, enabling real-time predictions and decisionmakingforusers.
Table -1: Systemmethodologyandworkflow
Phase Input
DataCollection Soildata,weather data,crophistory, diseaseimages
Data
Process Output
Datagathered fromdatasets Raw data
Preprocessing Rawdata Cleaning, normalization , feature engineering Processed dataset
ModelTraining Processeddataset ML models (Regression, CNN,SVM) Trained models
Deployment
Trainedmodels API integration, dashboard visualization Realtime predicti ons
2.5 AI and Financial Modeling
The system combines agricultural analytics with financial computations. AI models predict crop yield and detect diseases,whilefinancialmodelscalculateriskscores, expectedprofit,andreturnoninvestment(ROI).ESGfactors arealsoconsideredtopromotesustainableinvestments.
Table -2: AIModelsandFinancialComputation
Component Input
Cropsuitability
Method/Algo rithm
Soil, environment,irri gation Random forest,Decisi ontree,SVM
Diseasedetection
Yieldprediction
Risk&ESGanalysis
Profitcalculation
ROIcalculation
Output
Bestcrop predictio n
Leafimages CNN (Deep Learning) Disease classifica tion
Soil, crop, rainfall, temperature Regression models Yield estimatio n
Yield variability, marketdata,ESG
Yield, market price,cost
Weighted scoring model Risk Score
EP = Yield * price-cost Expected profit
Profit, investment ROI = (Profit/ investment) *100 ROI percenta ge
2.6 Performance Evaluation
Thesystemperformanceisevaluatedusingmetricssuchas accuracy, error rate, and ROI improvement. The results indicate high accuracy in crop prediction and disease detection,alongwithimprovedinvestmentdecision-making.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072
Accuracy(%)
95 ┤ ██████████ (96%)
94 ┤ ██████████ (94%)
92 ┤ ██████████ (92%)
90 ┤ ██████████ (90%)
85 ┤
Crop Disease Yield Matching Prediction Detection Forecast Engine
Chart-1: SystemPerformanceEvaluation
Thebarchartillustratestheperformanceofvariousmodules in the system. The disease detection model achieves the highestaccuracyduetotheuseofdeeplearningtechniques. Crop prediction and yield forecasting also demonstrate strongperformance,whilethestakeholdermatchingengine ensureseffectivealignmentbetweenfarmersandinvestors.
3. SYSTEM ARCHITECTURE
Thesystemfollowsa multi-layered architecture:
3.1 Frontend Layer
DevelopedusingReactJS
Providesuser-friendlydashboards
Separate interfaces for Farmer, Investor, Admin, andPartner
3.2 Backend Layer
BuiltusingSpringBoot
Handles business logic, APIs, authentication, and transactions
3.3 Database Layer
MongoDBforstoring:
o Userdata
o Projectdetails
o Investmentrecords
3.4 AI Layer
TensorFlow-basedmodelsfor:
o Cropprediction
o Diseasedetection
o Investmentrecommendation
4. SYSTEM MODULES
4.1 Farmer Module
Createandmanageagriculturalprojects
Inputsoiltype,croptype,andacreage
ReceiveAI-basedpredictions
Trackprojectprogressandprofits
4.2 Investor Module
Browseverifiedagriculturalprojects
ViewESGscoresandrisklevels
InvestusingSIPordirectfunding
MonitorROIandprofitdistribution
4.3 Admin Module
Verifyfarmerprojects
Approveorrejectwithreasons
Monitorplatformactivities
Manageusersandrevenue
4.4 Agri-Partner Module
Registerandprovideservices
Receiveassignedagriculturaltasks
Trackearningsandworkstatus
4.5 AI Chatbot Module
Providesreal-timeassistance
Explainsinvestmentstrategies Guidesfarmersoncropselection
5. METHODOLOGY
5.1 Data Collection
Soildata
Weatherdata
Cropdatasets
Diseaseimagedatasets
5.2 Data Preprocessing
Datacleaning
Normalization
Featureselection
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072
5.3 Model Development
CropPrediction
Regressionalgorithms(LinearRegression,Random Forest)
DiseaseDetection
ConvolutionalNeuralNetworks(CNN) InvestmentRecommendation
Rule-based+AIrecommendationsystems
5.4 System Integration
AImodelsintegratedwithbackendAPIs
Real-timepredictionsdeliveredtofrontend
6. RESULTS AND ANALYSIS
Thesystemdemonstrates:
Improved crop yield prediction accuracy (~85–90%)
EffectivediseasedetectionusingCNN
Better investment decision-making through ESG scoring
Increased transparency between farmers and investors
PerformanceMetrics
Accuracy
Precision
ROIimprovement
Riskreduction
7. ADVANTAGES
Data-drivendecision-making
Reducedagriculturalrisks
Improvedfinancialtransparency
Supportssustainabledevelopment
Easy-to-useinterface
8. LIMITATIONS
Requiresinternetconnectivity
Dependsondatasetquality
AImodelsrequiretrainingandupdates
9. CONCLUSIONS
Agri-InvestAIsuccessfullyintegratesArtificialIntelligence, agriculture,andfinancialinvestmentintoaunifiedplatform. Itenhancesproductivityforfarmersandenablesinvestorsto makeinformeddecisions.Thesystempromotessustainable
agriculture,reducesrisk,andimprovesprofitability,making itavaluablesolutionformodernagri-fintechchallenges.
ACKNOWLEDGEMENT
We express our sincere gratitude to our guide Prof.Chandana S.N for his valuable guidance, continuous support,andencouragementthroughoutthedevelopmentof this project. His insightful suggestions and technical expertisegreatlycontributedtothesuccessfulcompletionof thiswork.
We would also like to thank the Department of Computer ScienceandEngineering,CMRUniversity,forprovidingthe necessary resources and environment to carry out this project effectively. We extend our heartfelt thanks to all facultymembers,friends,andwell-wisherswhosupported us directly and indirectly during the development of this project. Finally, we are grateful to our families for their constant motivation and support, which helped us successfullycompletethisresearchwork.
REFERENCES
1.Lietal.(2021).DOI:10.1016/j.compag.2021.106189
2.Chenetal.(2022).DOI:10.1016/j.suscom.2022.100721
3.Gupta&Rao(2022).DOI:10.1002/ijfe.2460
4.Sharmaetal.(2023).DOI:10.1007/s42452-023-05412-0
5.Brown(2023).SpringerBooklink:10.1007/978-3-03124882-7
6. Shawon et al. (2024). ScienceDirect link: https://www.sciencedirect.com/science/article/pii/S27723 75524003228
7. Jabed et al. (2024). PMC link: https://pmc.ncbi.nlm.nih.gov/articles/PMC11667600/
8. Verma & Singh (2024). IEEE Xplore (link ascited): https://ieeexplore.ieee.org
9. Dhanaraj et al. (2025). Nature link: https://www.nature.com/articles/s41598-025-16014-4
10. Hernández et al. (2025). MDPI link: https://www.mdpi.com/2077-0472/15/23/2438