A study on horticultural development through ICT: A case study of e-governance in Janjgir Champa district in Chhattisgarh using classification technique

Introduction

E-governance has emerged as a critical facet of modern governance, enabling a shift from traditional bureaucratic systems toward digital, citizen-centric frameworks. Rapid advances in Information and Communication Technologies (ICT) have powered this transformation, strengthening the interactions across Government-to-Citizen (G2C), Government-to-Business (G2B), and Government-to-Government (G2G) domains. These models operate within legal and regulatory infrastructures to bolster efficiency, accountability, and transparency (Nahak, 2023; NIC Measures, 2022) [6] . Worldwide, governments are under escalating pressure to deliver public services that are both cost-effective and responsive. ICT has served as a catalyst by streamlining back-end administrative operations and enhancing front-end service delivery (United Nations e-Governance Survey, 2023) [12] . In India's rural areas, ICT-driven e-governance initiatives have improved service delivery at the community level, optimized data storage and retrieval, and propelled socio-economic development (Singh & Kumar, 2021; SmartGaon, 2022) [9,10] . Examples such as Bangla Sahayata Kendras and the UMANG platform have shown tangible gains: broader citizen access, substantial time savings, and reduced avenues for corruption. However, despite notable progress, persistent challenges limit widespread adoption. Key among these are limited public awareness, digital illiteracy, gaps in infrastructure, and linguistic diversity (Singh, 2024; "Accessibility of e-Governance in Rural India," 2018) [1,8] . Earlier projects like the Warana Wired Village and the Bhoomi land records initiative illustrated the potential of ICT for rural transformation but also underscored the difficulties of scalability and long-term sustainability (Warana evaluation, 2009; Bhoomi reports [13] ,

Madhya Pradesh, 2015) [2] . More recent research highlights the critical role of digital literacy programs, public-private collaboration, and robust policy frameworks to bridge the digital divide and ensure inclusivity (Moitra et al., 2021; Smart Village IoT studies, 2021) [5,11] . Against this backdrop, the present study examines how ICT can strengthen e-governance in rural horticultural settings. By identifying core constraints, exploring citizen perception differences, and applying classification methods, this research aims to offer insights that help policymakers build resilient, inclusive ICT-enabled governance systems.

Materials and methods

This study utilized a survey-based research design to examine the challenges associated with implementing ICT-related policies in the horticultural sector of Janjgir Champa District, Chhattisgarh. Conducted in 2025 across five administrative blocks, the survey ensured comprehensive coverage of the study area. A structured questionnaire was prepared to gather information on ICT-related issues, comprising 26 questions on demographic details, ICT usage, and perceived obstacles to egovernance adoption. It was divided into two sections: (i) general information on respondents' socio-economic characteristics and ICT access, and (ii) nine ICT-related constraints identified through prior literature and field consultations. These constraints were: (i) resource limitations, (ii) computer illiteracy, (iii) low citizen awareness, (iv) language barriers, (v) resistance to change, (vi) shortage of skilled personnel, (vii) equipment storage problems, (viii) perceived difficulty of use, and (ix) gender-related barriers. The survey included 120 respondents, primarily farmers, rural youth, and community members directly influenced by ICT initiatives in governance and horticulture. Data were gathered through face-to-face interviews and structured questionnaires, ensuring representation of individuals with different levels of digital literacy. The internal consistency of the instrument was tested using Cronbach's alpha, which ranged from 0.818 to 0.945 across the constraints, indicating high reliability. As per Nunnally (1978) [7] , scores above 0.70 denote acceptable reliability, those above 0.80 indicate good reliability, and those exceeding 0.90 reflect excellent reliability. Collected data were coded by the researcher and analyzed using statistical and machine learning tools, including Jamovi 2.6.44, Python 3.12, and R 4.3.1, which were employed for reliability testing, descriptive statistics, and application of classification algorithms. For classifying ICT-related challenges, three machine learning algorithms were applied: J-48, CART, and Random Forest. These were selected for their demonstrated effectiveness with categorical survey data and their interpretability in governance research. Model performance was evaluated using sensitivity, accuracy, specificity, and F1score. Among the three, Random Forest achieved the highest performance, showing superior accuracy and balanced predictive capability across all metrics.

Results Discussion Accuracy, Sensitivity, Specificity, and F1-Score of ICT based data for Horticulture Department

The CART method achieved an accuracy of 88.89%, sensitivity of 0.96, specificity of 0.97, and an F1-score of 0.85. The J48 method resulted in an accuracy of 86.11%, sensitivity of 0.92, specificity of 0.96, and an F1-score of 0.81. The Random Forest method delivered the best performance, with an accuracy of 94.44%, sensitivity of 0.95, specificity of 0.97, and an F1-score of 0.89. Based on these results, the Random Forest classifier outperformed the other methods and provided the most reliable classification results for the ICT data related to the Horticulture Department.

References

1. Accessibility of e-Governance in Rural India. J Rural Dev Stud. 2018;12(2):45-53.
2. Department of Land Records. Bhoomi reports, Madhya Pradesh: Annual Report on Land Records Modernization in Madhya Pradesh. Bhopal: Dept. of Land Records; 2015.
3. Breiman L. Random forests. Mach Learn. 2001;45(1):532.
4. Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and Regression Trees. Belmont: Wadsworth; 1984.
5. Moitra D, Sharma P, Kaur N. Strengthening rural egovernance through digital literacy programs: Evidence from Indian states. Int J E-Gov Res. 2021;17(3):88-104.
6. Nahak S. E-governance as a tool for citizen-centric administration in India. J Digit Gov. 2023;5(1):21-34.
7. Nunnally JC. Psychometric Theory. 2nd ed. New York: McGraw-Hill; 1978.
8. Singh R. Bridging the digital divide: E-governance challenges and opportunities in rural India. Indian J Public Adm. 2024;70(2):223-239.
9. Singh R, Kumar A. ICT-driven e-governance initiatives in India's rural sector: A review. Indian J Rural Dev. 2021;40(4):612-625.
10. SmartGaon. Smart village initiatives: Bridging digital gaps in India [Internet]. 2022 [cited 2025 Sep 22]. Available from: https://www.smartgaon.org/
11. Smart Village IoT studies. Smart village IoT applications for inclusive rural governance. Int J Smart Vill Res. 2021;3(2):55-68.
12. United Nations Department of Economic and Social Affairs. United Nations e-Government Survey 2023: Leveraging digital technology for sustainable development. New York: United Nations; 2023.
13. Ministry of Information Technology, Government of India. Warana Wired Village project: Evaluation report. New Delhi: MIT, GoI; 2009.